Compiler Design is quite important course from UGCNET /GATE point of view .This course clarifies different phases of language conversion.To have more insight refer http://tutorialfocus.net/

1. COMPILER DESIGN
INTRODUCTION
Richa Sharma 1

2. MAIN REFERENCE BOOK:
• COMPILERS – PRINCIPLES, TECHNIQUES AND TOOLS, SECOND EDITION BY
ALFRED V. AHO, RAVI SETHI, JEFFERY D. ULLMAN
• PRINCIPLES OF COMPILER DESIGN BY V R RAGHAVAN .
Richa Sharma 2

3. COMPILER
• IT’S A SOFTWARE UTILITY THAT TRANSLATED HIGH LANGUAGE CODE INTO TARGET LANGUAGE
CODE ,AS COMPUTER DOESN’T UNDERSTAND HIGH LANGUAGE.
DATA
OUTPUT
• IMPORTANT ROLE OF COMPILER IS TO REPORT THE ERROR IN THE SOURCE PROGRAM BY
TRANSLATING THE PROGRAM IN ONE GO.
• STRUCTURE OF THE COMPILER IS OFFLINE, MEANING THAT WE PRE-PROCESS THE PROGRAM FIRST
AND CREATES AND EXECUTABLE CODE AND THIS CODE CAN RUN ON DIFFERENT INPUTS OR DATA .
Richa Sharma 3
COMPILER
Source
Program
Target
Program

4. INTERPRETER
• IT’S ANOTHER SOFTWARE UTILITY THAT TRANSLATES HIGH LANGUAGE CODE INTO TARGET
LANGUAGE CODE LINE BY LINE UNLIKE COMPILER .
OUTPUT
• INTERPRETER IS ONLINE MODE, IN WHICH DATA AND SOURCE PROGRAM ARE EXECUTED
SIMULTANEOUSLY GIVING THE OUTPUT .NO PRE-PROCESSING IS DONE EARLIER.
Richa Sharma 4
COMPILER
Source Program
Data

5. EXAMPLES
MOST LANGUAGES ARE USUALLY THOUGHT OF AS USING EITHER
ONE OR THE OTHER:
• COMPILERS: FORTRAN, COBOL, C, C++, PASCAL, PL/1
• INTERPRETERS: LISP, SCHEME, BASIC, APL, PERL, PYTHON,
SMALLTALK
Richa Sharma 5

6. Preprocessor
Compiler
Assembler
Linker/Loader
Source program
Modified Source program
Target assembly program
Relocatable machine code
Target machine code
Library filesRicha Sharma
STEPS FOR LANGUAGE PROCESSING SYSTEM.

7. PHASES OF COMPILER
THE TRANSLATION OF INPUT FILE INTO TARGET CODE IS DIVIDED INTO 2 STAGES :
1. FRONT END (ANALYSIS): TRANSFORM SOURCE CODE INTO INTERMEDIATE CODE ALSO
CALLED INTERMEDIATE REPRESENTATION (IR) . IT’S A MACHINE-INDEPENDENT
REPRESENTATION.
2. BACK END (SYNTHESIS): IT TAKES IR AND GENERATES THE TARGET ASSEMBLY LANGUAGE
PROGRAM.
FRONT END BACK END
1) LEXICAL ANALYZER (SCANNING) 5) CODE OPTIMIZATION
2) SYNTAX ANALYZER (PARSING) 6) MACHINE CODE GENERATION
3) SEMANTIC ANALYZER
4) INTERMEDIATE CODE GENERATION
Richa Sharma 7

8. PHASES OF COMPILER
Richa Sharma 8
Symbol table Error handler
Front End
Back End

9. LEXICAL ANALYSIS/SCANNING
 READS THE STREAM OF CHARACTERS MAKING UP THE SOURCE PROGRAM AND GROUP THE
CHARACTERS INTO MEANINGFUL SEQUENCES CALLED LEXEMES.
 LEXEME ---- TOKEN
< TOKEN-NAME, ATTRIBUTE-VALUE>
 TOKEN NAME – IS AN ABSTRACT SYMBOL USED DURING SYNTAX ANALYSIS.
 ATTRIBUTE VALUE – POINTS TO AN ENTRY IN THE SYMBOL TABLE FOR THIS TOKEN.
POSITION = INITIAL + RATE * 60
<ID,1> < = > <ID,2> < +> <ID,3> < *> <60>
Richa Sharma 9

10. SYNTAX ANALYSIS/PARSING
• RECOGNIZES “SENTENCES” IN THE PROGRAM USING THE SYNTAX OF THE LANGUAGE
• CREATES TREE LIKE STRUCTURE FROM TOKENS (SYNTAX TREE)
• NODE REPRESENTS OPERATION
• CHILDREN REPRESENTS ARGUMENTS
• REPRESENTS THE SYNTACTIC STRUCTURE OF THE PROGRAM, HIDING A FEW DETAILS THAT ARE
IRRELEVENT TO LATER PHASES OF COMPILATION.
Richa Sharma 10

11. SEMANTIC ANALYSIS
• INFERS INFORMATION ABOUT THE PROGRAM USING THE SEMANTICS OF THE LANGUAGE
• USES SYNTAX TREE AND INFO. IN SYMBOL TABLE TO CHECK FOR SEMANTIC CONSISTENCY.
• GATHERS TYPE INFO. AND SAVES IT IN EITHER THE SYNTAX TREE OR SYMBOL TABLE FOR USE IN
ICG
• TYPE CHECKING – CHECKS THAT EACH OPERATOR HAS MATCHING OPERANDS. E.G ARRAY
INDEX SHOULD BE INTEGER.
• TYPE CONVERSIONS CALLED COERCIONS
• BINARY ARITHMETIC OPERATOR (INT OR FLOAT)
• IF 6+7.5, THEN CONVERT 6 TO 6.5
Richa Sharma 11

12. INTERMEDIATE CODE GENERATION
• GENERATES “ABSTRACT” CODE BASED ON THE SYNTACTIC STRUCTURE OF THE PROGRAM AND THE SEMANTIC
INFORMATION FROM PHASE 2.
 SYNTAX TREE ARE ALSO IR
 COMPILER MAY PRODUCE EXPLICIT IR
 IR HAS TWO PROPERTIES:
 EASY TO PRODUCE
 EASY TO TRANSLATE INTO TARGET MACHINE
 E.G IR – THREE ADDRESS CODE
 T1 = INTTOFLOAT(60)
 T2 = ID3 * T1
 T3 = ID2 + T3
 ID1 = T3
Richa Sharma 12

13. CODE OPTIMIZATION
• REFINES THE GENERATED CODE USING A SERIES OF OPTIMIZING TRANSFORMATIONS.
• Eg: REMOVING DEAD CODE .
REDUCING ITERATIONS AND LOOPS ETC..
• APPLY A SERIES OF TRANSFORMATIONS TO IMPROVE THE TIME AND SPACE EFFICIENCY OF THE
GENERATED CODE.
• PEEPHOLE OPTIMIZATIONS: GENERATE NEW INSTRUCTIONS BY COMBINING/EXPANDING ON
A SMALL NUMBER OF CONSECUTIVE INSTRUCTIONS.
• GLOBAL OPTIMIZATIONS: REORDER, REMOVE OR ADD INSTRUCTIONS TO CHANGE THE
STRUCTURE OF GENERATED CODE.
Richa Sharma 13

14. CODE GENEARTION
• MAP INSTRUCTIONS IN THE INTERMEDIATE CODE TO SPECIFIC MACHINE INSTRUCTIONS.
• SUPPORTS STANDARD OBJECT FILE FORMATS.
• GENERATES SUFFICIENT INFORMATION TO ENABLE SYMBOLIC DEBUGGING.
 IR -> CG -> TARGET LANGUAGE (E.G MACHINE CODE)
 REGISTERS AND MEMORY LOCATIONS ARE SELECTED FOR EACH VARIABLE USED BY THE PROGRAM.
 LDF R2, ID3
 MULF R2, R2, #60.0
 LDF R1, ID2
 ADDF R1, R1, R2
 STF ID1, R1
 R1,R2 – REGISTERS F – FLOATING POINT NUMBERS # - IMMEDIATE CONST.
Richa Sharma
14

15. SYMBOL TABLE
• SYMBOL TABLE – DATA STRUCTURE WITH A RECORD FOR EACH IDENTIFIER AND ITS ATTRIBUTES
• ALL THE PHASES ARE CONNECTED TO THE SYMBOL TABLE.
• ATTRIBUTES INCLUDE STORAGE ALLOCATION, TYPE, SCOPE, ETC
• ALL THE COMPILER PHASES INSERT AND MODIFY THE SYMBOL TABLE
Richa Sharma 15

16. Richa Sharma
16
OVERALL WORKING OF COMPILER PHASES

17. COMPILER CONSTRUCTION TOOLS
• DEVELOPER MAY USE MODERN SOFTWARE DEVELOPMENT ENVIRONMENT CONTAINING
TOOLS Ex: LANGUAGE EDITORS, VERSION MANAGERS, TEC.
• SOME SPECIAL TOOLS CAN ALSO BE USED. THESE TOOLS ARE THOSE WHICH HIDE THE DETAILS
OF GENERATION ALGORITHM AND PRODUCE COMPONENTS THAT CAN BE EASILY INTEGRATED
TO REMAINDER OF THE COMPILER.
• PARSER GENERATOR: TAKES GRAMMAR DESCRIPTION AND PRODUCES SYNTAX ANALYSER.
• SCANNER GENERATOR: TAKES REGULAR EXPRESSION AND PRODUCES LEXICAL ANALYSER.
• AUTOMATIC CODE- GENERATOR : TAKES INTERMEDIATE AND CONVERT TO MACHINE
LANGUAGE)
• DATA FLOW ANALYSIS ENGINES (FOR OPTIMIZATION)
• COMPILER CONSTRUCTION TOOLKIT.
Richa Sharma 17

18. APPLICATION OF COMPILER TECHNOLOGY
• IMPLEMENTATION OF HIGH LEVEL PROGRAMMING LANGUAGES
- CONCEPTS OF OOPS
• OPTIMIZATION FOR COMPUTER ARCHITECTURE
- PARALLELISM
- MEMORY HIERARCHIES OF MACHINES(REG, ARRAYS ETC)
• DESIGN OF NEW COMPUTER ARCHITECURES
- RISC
- CISC
- SIMD ETC
• PROGRAM TRANSLATIONS
• SOFTWARE PRODUCTIVITY TOOLS
Richa Sharma
18

19. Richa Sharma 19