关键词: 程序切片, 目标函数, 关键变量, 数据流分析, GIMPLE中间表示

Abstract: With the scale of program increasing and program complexity rising, program slicing through direct code analysis becomes much harder. In the existed approaches that use compiler optimization techniques for obtaining more accurate slices, compile-time information and existed compiler optimization are overlooked, program languages are not fully supported. To solve this problem, this paper analyzes the intermediate representations of GCC compiler, and proposes a program slicing technology based on critical variable dataflow analysis algorithm in GCC compiler. This algorithm collects dataflow information both within and between basic blocks through iteratively solving the dataflow equations, which is based on GIMPLE intermediate representation and processes a basic block in each time. The proposed technique can find the critical variables and critical statements relevant to target function, and reduces the scale of the sliced program. The results of experiment show that the proposed algorithm is feasible.

Key words: program slicing, target function, critical variants, data flow analysis, GIMPLE intermediate representation