Compiler Construction

COMP 451
Compiler: An Overview

1
 Today’s Adenda
▪ A basic intro in which we will see

2
 Today’s Adenda
▪ A basic intro in which we will see
— Types of compiler

3
 Today’s Adenda
▪ A basic intro in which we will see
— Types of compiler
— Major parts of compilation process

4
 Today’s Adenda
▪ A basic intro in which we will see
— Types of compiler
— Major parts of compilation process
— Other programs that work with compiler to execute a program

5
 Today’s Adenda
▪ A basic intro in which we will see
— Types of compiler
— Major parts of compilation process
— Other programs that work with compiler to execute a program
— Structure / phases of a compiler

6
 Today’s Adenda
▪ A basic intro in which we will see
— Types of compiler
— Major parts of compilation process
— Other programs that work with compiler to execute a program
— Structure / phases of a compiler
— A brief overview of functionality of all the phases of compiler using
an example.

7
 Compiler
▪ Compiler is a program that reads a program written in one language and
translates it into an equivalent program into another language.

8
 Compiler
▪ Compiler is a program that reads a program written in one language and
translates it into an equivalent program into another language.
▪ A compiler also reports errors present in the source program as part of
translation process.

9
 Compiler
▪ Compiler is a program that reads a program written in one language and
translates it into an equivalent program into another language.
▪ A compiler also reports errors present in the source program as part of
translation process.

10
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
▪Synthesis (Back End)

11
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)

12
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.

13
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.

14
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.
— Generates symbol table.

15
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.
— Generates symbol table.
▪Synthesis (Back End)

16
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.
— Generates symbol table.
▪Synthesis (Back End)
— Takes the intermediate representation as input.

17
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.
— Generates symbol table.
▪Synthesis (Back End)
— Takes the intermediate representation as input.
— Creates desired target program.

18
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.
— Generates symbol table.
▪Synthesis (Back End)
— Takes the intermediate representation as input.
— Creates desired target program.
▪Also called passes of a compilation process.

19
 Compilation Process
Major Parts of the Compilation Process
▪Analysis (Front End)
— Takes the input source program and breaks it into parts.
— Creates an intermediate representation of the source program.
— Generates symbol table.
▪Synthesis (Back End)
— Takes the intermediate representation as input.
— Creates desired target program.
▪Also called passes of a compilation process.

20
Phases of a Compiler

21
 Phases of a Compiler
Lexical Analyzer

22
 Phases of a Compiler
Lexical Analyzer
▪It reads the program and converts it into tokens.

23
 Phases of a Compiler
Lexical Analyzer
▪It reads the program and converts it into tokens.
▪Tokens are defined by regular expressions which are understood by the
lexical analyzer.

24
 Phases of a Compiler
Lexical Analyzer
▪It reads the program and converts it into tokens.
▪Tokens are defined by regular expressions which are understood by the
lexical analyzer.

25
 Phases of a Compiler
Syntax Analyzer

26
 Phases of a Compiler
Syntax Analyzer
▪Also called parser.

27
 Phases of a Compiler
Syntax Analyzer
▪Also called parser.
▪It takes all the tokens one by one and uses Context Free Grammar to
construct the parse tree.

28
 Phases of a Compiler
Syntax Analyzer
▪Also called parser.
▪It takes all the tokens one by one and uses Context Free Grammar to
construct the parse tree.

29
 Phases of a Compiler
Semantic Analyzer

30
 Phases of a Compiler
Semantic Analyzer
▪Verifies the parse tree, whether it’s meaningful or not.

31
 Phases of a Compiler
Semantic Analyzer
▪Verifies the parse tree, whether it’s meaningful or not.
▪Produces a verified/corrected/annotated parse tree.

32
 Phases of a Compiler
Semantic Analyzer
▪Verifies the parse tree, whether it’s meaningful or not.
▪Produces a verified/corrected/annotated parse tree.
▪It also performs type checking.

33
 Phases of a Compiler
Semantic Analyzer
▪Verifies the parse tree, whether it’s meaningful or not.
▪Produces a verified/corrected/annotated parse tree.
▪It also performs type checking.

34
 Phases of a Compiler
Intermediate Code Generator

35
 Phases of a Compiler
Intermediate Code Generator
▪Last phase of front end.

36
 Phases of a Compiler
Intermediate Code Generator
▪Last phase of front end.
▪An intermediate code is one which can be readily executed by machine.

37
 Phases of a Compiler
Intermediate Code Generator
▪Last phase of front end.
▪An intermediate code is one which can be readily executed by machine.
▪Three Address Code is a popular example of IC.

38
 Phases of a Compiler
Intermediate Code Generator
▪Last phase of front end.
▪An intermediate code is one which can be readily executed by machine.
▪Three Address Code is a popular example of IC.
▪Till intermediate code, it is same for every compiler out there, but after
that, it depends on the platform.

39
 Phases of a Compiler
Intermediate Code Generator
▪Last phase of front end.
▪An intermediate code is one which can be readily executed by machine.
▪Three Address Code is a popular example of IC.
▪Till intermediate code, it is same for every compiler out there, but after
that, it depends on the platform.
▪To build a new compiler we don’t need to build it from scratch.

40
 Phases of a Compiler
Intermediate Code Generator
▪Last phase of front end.
▪An intermediate code is one which can be readily executed by machine.
▪Three Address Code is a popular example of IC.
▪Till intermediate code, it is same for every compiler out there, but after
that, it depends on the platform.
▪To build a new compiler we don’t need to build it from scratch.
▪We can take the intermediate code from the already existing compiler and
build the last two parts.

41
 Phases of a Compiler
Code Optimizer

42
 Phases of a Compiler
Code Optimizer
▪It transforms the code so that it consumes fewer resources and produces
more speed.

43
 Phases of a Compiler
Code Optimizer
▪It transforms the code so that it consumes fewer resources and produces
more speed.
▪The meaning of the code being transformed is not altered. 

44
 Phases of a Compiler
Code Optimizer
▪It transforms the code so that it consumes fewer resources and produces
more speed.
▪The meaning of the code being transformed is not altered. 

Target Code Generator

45
 Phases of a Compiler
Code Optimizer
▪It transforms the code so that it consumes fewer resources and produces
more speed.
▪The meaning of the code being transformed is not altered. 

Target Code Generator

▪This is the final stage of compilation.

46
 Phases of a Compiler
Code Optimizer
▪It transforms the code so that it consumes fewer resources and produces
more speed.
▪The meaning of the code being transformed is not altered. 

Target Code Generator

▪This is the final stage of compilation.
▪The main purpose of Target Code generator is to write a code that the
machine can understand.

47
 Phases of a Compiler
Code Optimizer
▪It transforms the code so that it consumes fewer resources and produces
more speed.
▪The meaning of the code being transformed is not altered. 

Target Code Generator

▪This is the final stage of compilation.
▪The main purpose of Target Code generator is to write a code that the
machine can understand.
▪The output is dependent on the type of assembler.

48
 Symbol Table
▪ It stores information
— about scope

49
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.

50
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.

51
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.

52
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.

53
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.

54
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.
▪ Items stored in symbol table:

55
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.
▪ Items stored in symbol table:
— Variable names and constants

56
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.
▪ Items stored in symbol table:
— Variable names and constants
— Procedure and function names

57
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.
▪ Items stored in symbol table:
— Variable names and constants
— Procedure and function names
— Literal constants and strings

58
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.
▪ Items stored in symbol table:
— Variable names and constants
— Procedure and function names
— Literal constants and strings
— Compiler generated temporaries

59
 Symbol Table
▪ It stores information
— about scope
— about instances of various entities such as variable and function
names, classes, objects, etc.
▪ It is populated in lexical and syntax analysis phases.
▪ The information is collected by the analysis phases of compiler and is
used by synthesis phases of compiler to generate code.
▪ It is used by compiler to achieve compile time efficiency.
▪ Each entry in symbol table is associated with attributes that support
compiler in different phases.
▪ Items stored in symbol table:
— Variable names and constants
— Procedure and function names
— Literal constants and strings
— Compiler generated temporaries
— Labels in source languages

60
 Broad Categories of Compilers

▪Single Pass Compiler

▪Multi Pass Compiler

61
Single Pass Compiler

62
 Single Pass Compiler

Single pass compiler

63
 Single Pass Compiler

Single pass compiler

•passes through the part of each
compilation unit exactly once.

64
 Single Pass Compiler

Single pass compiler

•passes through the part of each
compilation unit exactly once.
•faster and smaller than the multi
pass compiler.

65
 Single Pass Compiler

Single pass compiler

•passes through the part of each
compilation unit exactly once.
•faster and smaller than the multi
pass compiler.
•less efficient in comparison with
multi-pass compiler.

66
 Single Pass Compiler

Single pass compiler

•passes through the part of each
compilation unit exactly once.
•faster and smaller than the multi
pass compiler.
•less efficient in comparison with
multi-pass compiler.
•processes the input exactly once,
so going directly from lexical
analysis to code generator, and
then going back for the next
read.

67
 Single Pass Compiler

Single pass compiler

•passes through the part of each
compilation unit exactly once.
•faster and smaller than the multi
pass compiler.
•less efficient in comparison with
multi-pass compiler.
•processes the input exactly once,
so going directly from lexical
analysis to code generator, and
then going back for the next
read.
•can’t backup and re-process, so
grammar should be simplified.

68
 Single Pass Compiler

Single pass compiler

•passes through the part of each
compilation unit exactly once.
•faster and smaller than the multi
pass compiler.
•less efficient in comparison with
multi-pass compiler.
•processes the input exactly once,
so going directly from lexical
analysis to code generator, and
then going back for the next
read.
•can’t backup and re-process, so
grammar should be simplified.
•Examples may include command
interpreters such as bash/sh/tcsh

69
Two Pass Compiler

70
 Two Pass Compiler
Two pass compiler

71
 Two Pass Compiler
Two pass compiler
•processes the source code of a
program multiple times.

72
 Two Pass Compiler
Two pass compiler
•processes the source code of a
program multiple times.
•In multi-pass compiler, we divide
phases in two passes.

73
 Two Pass Compiler
Two pass compiler
•processes the source code of a
program multiple times.
•In multi-pass compiler, we divide
phases in two passes.
•First pass is called front end, and
is platform independent.

74
 Two Pass Compiler
Two pass compiler
•processes the source code of a
program multiple times.
•In multi-pass compiler, we divide
phases in two passes.
•First pass is called front end, and
is platform independent.
•Second pass is called back end
and is platform dependent.

75
Front End / Back End Division

76
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.

77
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.

78
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end

79
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,

80
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,
— determines its constituent parts, and

81
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,
— determines its constituent parts, and
— constructs an intermediate representation of the program.

82
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,
— determines its constituent parts, and
— constructs an intermediate representation of the program.
▪ Typically the front end is independent of the target language.

83
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,
— determines its constituent parts, and
— constructs an intermediate representation of the program.
▪ Typically the front end is independent of the target language.
▪ The back end

84
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,
— determines its constituent parts, and
— constructs an intermediate representation of the program.
▪ Typically the front end is independent of the target language.
▪ The back end
— synthesizes the target program from the intermediate representation
produced by the front end.

85
 Front End / Back End Division
▪ As discussed earlier, modern compilers contain two (large) parts, each of
which is often subdivided further.
▪ These two parts are the front end, and the back end.
▪ The front end
— analyzes the source program,
— determines its constituent parts, and
— constructs an intermediate representation of the program.
▪ Typically the front end is independent of the target language.
▪ The back end
— synthesizes the target program from the intermediate representation
produced by the front end.
▪ Typically the back end is independent of the source language.

86
 Front End / Back End Division
▪ This front/back division very much reduces the work for a compiling
system that can handle several (N) source languages and several (M)
target languages.

87
 Front End / Back End Division
▪ This front/back division very much reduces the work for a compiling
system that can handle several (N) source languages and several (M)
target languages.
▪ Instead of NM compilers, we need N front ends and M back ends.

88
 Front End / Back End Division
▪ This front/back division very much reduces the work for a compiling
system that can handle several (N) source languages and several (M)
target languages.
▪ Instead of NM compilers, we need N front ends and M back ends.
▪ For gcc (originally abbreviating "Gnu C Compiler", but now abbreviating
"Gnu Compiler Collection"), N=7 and M~30 so the savings are
considerable.

89
Front End / Back End Division

90
 Front End / Back End Division

• Compiler for more than

one programming
languages but same
machine.

91
 Front End / Back End Division

• Compiler for more than

one programming
languages but same
machine.
• For each programming
language there is
requirement of making
Front end/first pass for
each of them and only
one Back end/second
pass as shown.

92
Front End / Back End Division

93
 Front End / Back End Division

• compiler for same

programming language
but different
machines/systems.

94
 Front End / Back End Division

• compiler for same

programming language
but different
machines/systems.
• In this case we make
different Back ends for
different machines and
make only one Front
end for same
programming language
as shown.

95
Front End / Back End Division

96
 Programs which help
Compiler accomplish its task

97
 Cousins of Compilers
Cousins of Compiler

98
 Cousins of Compilers
Cousins of Compiler
▪These are different programs that work with compiler to generate machine
code.

99
 Cousins of Compilers
Cousins of Compiler
▪These are different programs that work with compiler to generate machine
code.
▪Other than compiler, there are three programs that work together to do the
task

100
 Cousins of Compilers
Cousins of Compiler
▪These are different programs that work with compiler to generate machine
code.
▪Other than compiler, there are three programs that work together to do the
task
— Pre processor

101
 Cousins of Compilers
Cousins of Compiler
▪These are different programs that work with compiler to generate machine
code.
▪Other than compiler, there are three programs that work together to do the
task
— Pre processor
— Assembler

102
 Cousins of Compilers
Cousins of Compiler
▪These are different programs that work with compiler to generate machine
code.
▪Other than compiler, there are three programs that work together to do the
task
— Pre processor
— Assembler
— Loader/Linker

103
Cousins of Compilers

104
 Cousins of Compilers
Pre processor

105
 Cousins of Compilers
Pre processor
▪Macro expansion

106
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion

107
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.

108
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.
▪Strip off comments

109
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.
▪Strip off comments
Compiler

110
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.
▪Strip off comments
Compiler
▪Checks for syntax and semantic errors.

111
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.
▪Strip off comments
Compiler
▪Checks for syntax and semantic errors.
▪If no error, the expanded code is converted into assembly code specific for
the processor architecture.

112
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.
▪Strip off comments
Compiler
▪Checks for syntax and semantic errors.
▪If no error, the expanded code is converted into assembly code specific for
the processor architecture.
Assembler

113
 Cousins of Compilers
Pre processor
▪Macro expansion
▪File inclusion
▪Provides a modified source program.
▪Strip off comments
Compiler
▪Checks for syntax and semantic errors.
▪If no error, the expanded code is converted into assembly code specific for
the processor architecture.
Assembler
▪Produces a re-locatable machine code.

114
 Cousins of Compilers
Linker

115
 Cousins of Compilers
Linker
▪Combines object files and library routines required to run the complete
source program.

116
 Cousins of Compilers
Linker
▪Combines object files and library routines required to run the complete
source program.
▪It then hands over the stuff to the loader.
Loader

117
 Cousins of Compilers
Linker
▪Combines object files and library routines required to run the complete
source program.
▪It then hands over the stuff to the loader.
Loader
▪Takes the entire program (the one provided by the assembler and the linker)
to main memory for execution.

118
 Cousins of Compilers: Practical Review
▪ Write a simple program in C on *NIX.

119
 Cousins of Compilers: Practical Review
▪ Write a simple program in C on *NIX.
▪ Compile it using
$ gcc –save-temps prg.c

120
 Cousins of Compilers: Practical Review
▪ Write a simple program in C on *NIX.
▪ Compile it using
$ gcc –save-temps prg.c
▪ This will produce four files prg.i, prg.s, prg.o, and a.out

121
 Cousins of Compilers: Practical Review
▪ Write a simple program in C on *NIX.
▪ Compile it using
$ gcc –save-temps prg.c
▪ This will produce four files prg.i, prg.s, prg.o, and a.out
▪ Go through these files to understand the working of cousins of compilers
discussed in previous slides.

122
Linking: Static Vs Dynamic

123
 Linking: Static Vs Dynamic
▪ Linking is of two types.

124
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
— Dynamic linking

125
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.

126
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.
— Dynamic linking
• Code placement is delayed until load time / run time.

127
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.
— Dynamic linking
• Code placement is delayed until load time / run time.
▪ To illustrate the concept, try compiling the program prg.c as follows

128
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.
— Dynamic linking
• Code placement is delayed until load time / run time.
▪ To illustrate the concept, try compiling any C program as follows
$ gcc –static prg.c –o prgstatic

129
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.
— Dynamic linking
• Code placement is delayed until load time / run time.
▪ To illustrate the concept, try compiling any C program as follows
$ gcc –static prg.c –o prgstatic
And
$ gcc prg.c –o prgdyn

130
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.
— Dynamic linking
• Code placement is delayed until load time / run time.
▪ To illustrate the concept, try compiling any C program as follows
$ gcc –static prg.c –o prgstatic
And
$ gcc prg.c –o prgdyn

▪ Now assuming both prgstatic and prgdyn are in the PWD

$ ls –li

131
 Linking: Static Vs Dynamic
▪ Linking is of two types.
— Static linking
• Code of external function/s is placed in the executable.
— Dynamic linking
• Code placement is delayed until load time / run time.
▪ To illustrate the concept, try compiling any C program as follows
$ gcc –static prg.c –o prgstatic
And
$ gcc prg.c –o prgdyn

▪ Now assuming both prgstatic and prgdyn are in the PWD

$ ls –li
▪ Look at the size of both executables.

132
Coming Back to Phases of a
Compiler

Understanding with an Example

133
 Lexical Analysis (Scanning)
Source Text into Tokens

134
 Lexical Analysis (Scanning)
Source Text into Tokens
▪Reading the input text (character by character) and grouping individual
characters into tokens such as identifiers, integers, reserved words, and
delimiters.

135
 Lexical Analysis (Scanning)
Source Text into Tokens
▪Reading the input text (character by character) and grouping individual
characters into tokens such as identifiers, integers, reserved words, and
delimiters.
▪How to describe these tokens?

136
 Lexical Analysis (Scanning)
Source Text into Tokens
▪Reading the input text (character by character) and grouping individual
characters into tokens such as identifiers, integers, reserved words, and
delimiters.
▪How to describe these tokens?
▪Regular expression notation is an effective approach to describing tokens.

137
 Lexical Analysis (Scanning)
Lexemes and Tokens

138
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,

139
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,
▪Tokens are the output of the lexical analyzer.

140
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,
▪Tokens are the output of the lexical analyzer.
▪For example, any one of the following C statements

141
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,
▪Tokens are the output of the lexical analyzer.
▪For example, any one of the following C statements
x3 = y + 3;
x3=y + 3 ;
x3=y+3 ;

142
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,
▪Tokens are the output of the lexical analyzer.
▪For example, any one of the following C statements
x3 = y + 3;
x3=y + 3 ;
x3=y+3 ;
would be grouped into the lexemes x3, =, y, +, 3, and ;

143
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,
▪Tokens are the output of the lexical analyzer.
▪For example, any one of the following C statements
x3 = y + 3;
x3=y + 3 ;
x3=y+3 ;
would be grouped into the lexemes x3, =, y, +, 3, and ;

but the following would result in

x 3 = y + 3;

144
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The character stream input is grouped into units called lexemes, which are
then mapped into tokens,
▪Tokens are the output of the lexical analyzer.
▪For example, any one of the following C statements
x3 = y + 3;
x3=y + 3 ;
x3=y+3 ;
would be grouped into the lexemes x3, =, y, +, 3, and ;

but the following would result in

x 3 = y + 3;

x, 3, = y, +, 3 and ;

145
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪For example, any one of the following C statements
x3 = y + 3;
x3 = y + 3 ;
x3 =y+3 ;
but not
x 3 = y + 3;
would be grouped into the lexemes x3, =, y, +, 3, and ;

146
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪For example, any one of the following C statements
x3 = y + 3;
x3 = y + 3 ;
x3 =y+3 ;
but not
x 3 = y + 3;
would be grouped into the lexemes x3, =, y, +, 3, and ;
▪A token is a <token-name, attribute-value> pair. For example

147
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪For example, any one of the following C statements
x3 = y + 3;
x3 = y + 3 ;
x3 =y+3 ;
but not
x 3 = y + 3;
would be grouped into the lexemes x3, =, y, +, 3, and ;
▪A token is a <token-name, attribute-value> pair. For example
— The lexeme x3 would be mapped to a token such as <id,1>.

148
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪For example, any one of the following C statements
x3 = y + 3;
x3 = y + 3 ;
x3 =y+3 ;
but not
x 3 = y + 3;
would be grouped into the lexemes x3, =, y, +, 3, and ;
▪A token is a <token-name, attribute-value> pair. For example
— The lexeme x3 would be mapped to a token such as <id,1>.
— The name id is short for identifier.

149
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪For example, any one of the following C statements
x3 = y + 3;
x3 = y + 3 ;
x3 =y+3 ;
but not
x 3 = y + 3;
would be grouped into the lexemes x3, =, y, +, 3, and ;
▪A token is a <token-name, attribute-value> pair. For example
— The lexeme x3 would be mapped to a token such as <id,1>.
— The name id is short for identifier.
— The value 1 is the index of the entry for x3 in the symbol table
produced by the compiler.

150
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme =

151
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme =
— would be mapped to the token <=>.

152
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme =
— would be mapped to the token <=>.
— In reality it is probably mapped to a pair, whose second component is
ignored.

153
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme =
— would be mapped to the token <=>.
— In reality it is probably mapped to a pair, whose second component is
ignored.
— The point is that there are many different identifiers so we need the
second component, but there is only one assignment symbol =.

154
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme =
— would be mapped to the token <=>.
— In reality it is probably mapped to a pair, whose second component is
ignored.
— The point is that there are many different identifiers so we need the
second component, but there is only one assignment symbol =.
▪The lexeme y is mapped to the token <id,2>

155
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme =
— would be mapped to the token <=>.
— In reality it is probably mapped to a pair, whose second component is
ignored.
— The point is that there are many different identifiers so we need the
second component, but there is only one assignment symbol =.
▪The lexeme y is mapped to the token <id,2>
▪The lexeme + is mapped to the token <+>.

156
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.

157
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>

158
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>
▪Note that non-significant blanks are normally removed during scanning.

159
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>
▪Note that non-significant blanks are normally removed during scanning.
▪ In C, most blanks are non-significant.

160
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>
▪Note that non-significant blanks are normally removed during scanning.
▪ In C, most blanks are non-significant.
▪That does not mean the blanks are unnecessary. Consider

161
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>
▪Note that non-significant blanks are normally removed during scanning.
▪ In C, most blanks are non-significant.
▪That does not mean the blanks are unnecessary. Consider
int x; and
intx;

162
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>
▪Note that non-significant blanks are normally removed during scanning.
▪ In C, most blanks are non-significant.
▪That does not mean the blanks are unnecessary. Consider
int x; and
intx;
▪The blank between int and x is clearly necessary, but it does not become
part of any token.

163
 Lexical Analysis (Scanning)
Lexemes and Tokens
▪The lexeme ‘3’ is somewhat interesting and will be discussed later.
▪The lexeme ‘;’ is mapped to the token <;>
▪Note that non-significant blanks are normally removed during scanning.
▪ In C, most blanks are non-significant.
▪That does not mean the blanks are unnecessary. Consider
int x; and
intx;
▪The blank between int and x is clearly necessary, but it does not become
part of any token.
▪Blanks inside strings are an exception, they are strictly considered as part
of the token.

164
 Syntax Analysis (Parsing)
▪ The parser is based on a formal syntax specification such as a CFGs.

165
 Syntax Analysis (Parsing)
▪ The parser is based on a formal syntax specification such as a CFGs.
▪ What is a CFG?

166
 Syntax Analysis (Parsing)
▪ The parser is based on a formal syntax specification such as a CFGs.
▪ What is a CFG?
— Context-free Grammars allow a non-terminal to be replaced by a
corresponding production rule whenever it appears in a derivation
process.

167
 Syntax Analysis (Parsing)
▪ The parser is based on a formal syntax specification such as a CFGs.
▪ What is a CFG?
— Context-free Grammars allow a non-terminal to be replaced by a
corresponding production rule whenever it appears in a derivation
process.
— The replacement occurs irrespective of what lies before or after the
non-terminal. 

168
 Syntax Analysis (Parsing)
▪ The parser is based on a formal syntax specification such as a CFGs.
▪ What is a CFG?
— Context-free Grammars allow a non-terminal to be replaced by a
corresponding production rule whenever it appears in a derivation
process.
— The replacement occurs irrespective of what lies before or after the
non-terminal. 
▪ And what is a CSG?

169
 Syntax Analysis (Parsing)
▪ The parser is based on a formal syntax specification such as a CFGs.
▪ What is a CFG?
— Context-free Grammars allow a non-terminal to be replaced by a
corresponding production rule whenever it appears in a derivation
process.
— The replacement occurs irrespective of what lies before or after the
non-terminal. 
▪ And what is a CSG?
— Context-sensitive Grammars allow replacement of a non-terminal
based on what lies before or after the non-terminal.

170
 Syntax Analysis (Parsing)
▪ Consider the rule 

171
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

172
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

▪ What this says is "wherever you find 𝐴, you can replace it with 0𝐴1".

173
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

▪ What this says is "wherever you find 𝐴, you can replace it with 0𝐴1".
▪ Now, consider the rule

174
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

▪ What this says is "wherever you find 𝐴, you can replace it with 0𝐴1".
▪ Now, consider the rule

𝐶𝐴𝐵→𝐶0𝐴1𝐵

175
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

▪ What this says is "wherever you find 𝐴, you can replace it with 0𝐴1".
▪ Now, consider the rule

𝐶𝐴𝐵→𝐶0𝐴1𝐵

▪ This says "You can replace 𝐴 with 0𝐴1 only if it is preceded by 𝐶 and

followed by 𝐵”.

176
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

▪ What this says is "wherever you find 𝐴, you can replace it with 0𝐴1".
▪ Now, consider the rule

𝐶𝐴𝐵→𝐶0𝐴1𝐵

▪ This says "You can replace 𝐴 with 0𝐴1 only if it is preceded by 𝐶 and

followed by 𝐵”.
▪ Here, it imposes a condition on when 𝐴 can be replaced with 0𝐴1.

177
 Syntax Analysis (Parsing)
▪ Consider the rule 

𝐴→0𝐴1

▪ What this says is "wherever you find 𝐴, you can replace it with 0𝐴1".
▪ Now, consider the rule

𝐶𝐴𝐵→𝐶0𝐴1𝐵

▪ This says "You can replace 𝐴 with 0𝐴1 only if it is preceded by 𝐶 and

followed by 𝐵”.
▪ Here, it imposes a condition on when 𝐴 can be replaced with 0𝐴1.
▪ You can apply this rule only if 𝐴 appears in this particular context.

178
 Syntax Analysis (Parsing)
▪ Parsing involves a further grouping in which tokens are grouped into
grammatical phrases, which are often represented in a parse tree.

179
 Syntax Analysis (Parsing)
▪ Parsing involves a further grouping in which tokens are grouped into
grammatical phrases, which are often represented in a parse tree.
▪ For example
x3 = y + 3;

180
 Syntax Analysis (Parsing)
▪ Parsing involves a further grouping in which tokens are grouped into
grammatical phrases, which are often represented in a parse tree.
▪ For example
x3 = y + 3;
would be parsed into a tree.

181
 Syntax Analysis (Parsing)
▪ Parsing involves a further grouping in which tokens are grouped into
grammatical phrases, which are often represented in a parse tree.
▪ For example
x3 = y + 3;
would be parsed into a tree.
▪ This parsing would result from a grammar containing rules such as
asst-stmt → id = expr ;
expr → number | id | expr + expr

182
 Syntax Analysis (Parsing)
▪ Parsing involves a further grouping in which tokens are grouped into
grammatical phrases, which are often represented in a parse tree.
▪ For example
x3 = y + 3;
would be parsed into a tree.
▪ This parsing would result from a grammar containing rules such as
asst-stmt → id = expr ;
expr → number | id | expr + expr
▪ Note the recursive definition of expression (expr).

183
Syntax Analysis (Parsing)

184
 Semantic Analysis
▪ Also called the Type Checker.

185
 Semantic Analysis
▪ Also called the Type Checker.
▪ Type checking is purely dependent on the semantic rules of the source
language. It is independent of the compiler’s target.

186
 Semantic Analysis
▪ Also called the Type Checker.
▪ Type checking is purely dependent on the semantic rules of the source
language. It is independent of the compiler’s target.
▪ The semantics of a language provide a set of rules that specify which
syntactically legal programs are actually valid.

187
 Semantic Analysis
▪ Also called the Type Checker.
▪ Type checking is purely dependent on the semantic rules of the source
language. It is independent of the compiler’s target.
▪ The semantics of a language provide a set of rules that specify which
syntactically legal programs are actually valid.
▪ Such rules typically require that
— all identifiers be declared,
— operators and operands be type-compatible, and
— procedures be called with the proper number of parameters.

188
 Intermediate Code Generation
▪ Many compilers internally generate intermediate code for an "idealized
machine".

189
 Intermediate Code Generation
▪ Many compilers internally generate intermediate code for an "idealized
machine".
▪ For example, the intermediate code generated would assume that the
target has an unlimited number of registers and that any register can be
used for any operation.

190
 Intermediate Code Generation
▪ Many compilers internally generate intermediate code for an "idealized
machine".
▪ For example, the intermediate code generated would assume that the
target has an unlimited number of registers and that any register can be
used for any operation.
▪ With these assumptions of a machine with an unlimited number of
registers and instructions with three operands, one generates
"three-address code" by walking the semantic tree.

191
 Code Optimazation
▪ This is a very serious subject, one that we will not really do justice to in
this introductory course. Some optimizations are fairly easy to see.
▪ In addition to optimizations performed on the intermediate code, further
optimizations can be performed on the machine code by the
machine-dependent back end.

192
 Code Generation
▪ Modern processors have only a limited number of register.
▪ Some processors (e.g., the MIPS architecture) use three-address
instructions.
▪ Other processors permit only two addresses; the result overwrites one of
the sources.
▪ Using three-address instructions restricted to registers (except for load
and store instructions, which naturally must also reference memory),
code something like the following would be produced for our example,
after first assigning memory locations to id1 and id2.
LD R1, id2
ADDF R1, R1, #3.0 // add float
RTOI R2, R1 // real to int
ST id1, R2

193