Design and Implementation of Bytecode-Based Java Slicing System

1
Design and Implementation of Bytecode-Based Java
Slicing System

• Fumiaki Umemori, Kenji Konda,
• Reishi Yokomori, Katsuro Inoue
• Osaka University, Japan

2
Background

• Software systems are becoming large and complex.
• Developers spend a large amount of time to test
  and debug their systems.
• Various techniques for improving debugging
  efficiency have been proposed
• Program Slicing

3
Dependence-Cache (DC) Slicing

• Dependence-Cache Slicing
• Intermediate slicing between static slicing and
  dynamic slicing
• DD(Data Dependence), CD(Control Dependence)
• Advantage
• Better analysis precision than static slice
• Smaller analysis cost than dynamic slice

Static slice DC slice Dynamic slice
DD analysis Static Dynamic Dynamic
CD analysis Static Static Dynamic
Target Source code Source code Execution trace
Ashida, Y., Ohata, F. and Inoue, K. Slicing
Methods Using Static and Dynamic Information,
Proceedings of the 6th Asia Pacific Software
Engineering Conference, 344-350, 1999.
4
Dynamic Data Dependence Analysis for DC Slicing

• Key idea Cache
• On program execution, we have only to trace the
  most-recently defined statement for each variable
  using cache.
• Cache(v) statement that defined the value of v
  most-recently.
• Operations for caches
• Before program execution,
• For each variable v, Cache(v) ? ?.
• On program execution,
• For each statement s,
• when v is defined, Cache(v) ? s.
• when v is referred to, we extract dynamic data
  dependence
• from Cache(v) to s about v.

5
Slicing for Java

• Java programs have many elements determined at
  runtime
• Dynamic DD analysis is necessary
• Proposed Method
• Analyze dynamic data dependence using Java
  Virtual Machine
• Analysis is based on bytecode
• Mapping bytecode onto source code

DC slicing is effective for Java Slicing System
6
Data Dependence Analysis Based on Java Virtual
Machine
Java Virtual Machine
s1 iconst_3
local variable
stack
s2 iconst_5
s3 iadd
8
0
s4 istore_0
corresponding cache
data dependence
s4
0
7
System Implementation

• Bytecode-Based Java DC-Slicing System
• (U1) Compile unit
• Extended Java compiler to create cross reference
  table
• (U2) Control Dependence Analysis unit
• Extract CD relation statically
• (U3) Execution unit
• Extended Java Virtual Machine(JVM) to extract DD
  relation dynamically
• (U4) Slicing unit
• Construct Program Dependence Graph(PDG) and
  Compute Slice

8
Overview of Slice Calculation
Java Compile
Slice Criterion
User Input
Tool
Slicer
Tool Output
CD analysis tool
static control dependence
dynamic data dependence
Java Virtual Machine
9
Conclusions and Future Works

• Bytecode based Java Slicing System
• Cross Reference Table
• Static Analysis for Control Dependence
• Dynamic Analysis for Data Dependence
• Slicer
• Future Works
• Cross Reference Table for optimized bytecode
• Improve Data Dependence analysis time

10
END
Thanks ! Question ? (But, please speak slowly !)