PF-Miner

Journal of Systems and Software(2016)

引用 2|浏览80
暂无评分
摘要
Proposed a method for mining paired function on the source code of drivers.Sequence dependency rules based on paired functions are also proposed.A theory model base on paired function has been proposed for violation detection.PF-Miner mined 546 paired functions, 28"49"(not 28) sequence dependency rules.15 bugs have been confirmed for the top 51 violations based on mining results. Generally, drivers have many errors to handle, and the functions called in the normal execution paths and error handling paths are in pairs, which are named as paired functions. However, some developers do not handle the errors completely as they forget or unaware of releasing the acquired resources, thus memory leaks and other potential problems can be easily introduced into the system. Therefore, it is highly valuable to automatically extract paired functions and detect violations for programmers.This paper proposes an efficient tool named PF-Miner, which can automatically extract paired functions and detect violations between normal execution paths and error handling paths from the source code of drivers with the data mining and statistical methods. We have evaluated PF-Miner on two different versions of Android kernel 2.6.39 and 3.10.0, and 81 bugs reported by PF-Miner in 2.6.39 have been fixed before the latest version 3.10.0. PF-Miner only needs about 150 s to analyze the source code of 3.10.0, and 983 violations have been detected from 546 paired functions which have been extracted. We have reported the top 51 violations as potential bugs to the developers, and 15 bugs have been confirmed so far.
更多
查看译文
关键词
Error path checking,Paired function mining,Violations detection
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要