An Evaluation of Slicing Algorithms for Concurrent Programs

Abstract

Program slicing is a program-reduction technique for extracting statements that may influence other statements. While there exist efficient algorithms to slice sequential programs precisely, there are only two algorithms for precise slicing of concurrent interprocedural programs with recursive procedures. We implemented both algorithms for Java, applied several new optimizations and examined their precision and runtime behavior. We compared these results with two further algorithms which trade precision for speed. We show that one algorithm may produce incorrect slices and that precise slicing of concurrent programs is very expensive in terms of computation time.

Static path conditions for Java

“…as the backward data slice contains only the allocation in line 9, so this condition is more precise than the basic formula (11). Note that ( * * ) was already more precise than ( * ), while ( * * * ) now collapses to a simple condition without type tests "x < 2", which is more precise than the "x < 3" in ( * ).…”

“…Alternatively, only possible program executions, so-called threaded witnesses, are considered valid paths for path condition computation, which requires more expensive slicing and chopping algorithms. An evaluation of precise concurrent slicing algorithms can be found in Giffhorn and Hammer's work [11].…”

Slicing Concurrent Programs Based on Program Reachability Graphs

“…Program slicing has been successfully applied in many software engineering activities, such as program understanding, debugging, testing, maintenance, measurement, reengineering [2]. Unfortunately, most approaches and tools for sequential programs cannot be directly applicable to concurrent programs [3][4][5][6][7][8][9][10][11][12][13][14].…”

“…Thus, we try to compute a slice that is as precise as possible. Up to now, researchers have proposed several slicing algorithms by additionally judging whether dependence statement sequences meet the requirement of execution order or not when traversing program dependence graphs [4][5][6][7][8][9][10][11][12][13][14]. Although they may remove considerable redundant statements and improve the precision, these methods have not handled the intransitive dependence problem inherently.…”

“…Although they may remove considerable redundant statements and improve the precision, these methods have not handled the intransitive dependence problem inherently. When there is a redefinition of a shared variable by another thread, some dependence statement sequences may be still intransitive despite of meeting the requirement of execution order [13][14]. If we settle the problem, more precise slice will be obtained.…”