类Unix文件系统中TOCTTOU缺陷的静态分析方法  被引量:2

Static Analysis of TOCTTOU Vulnerabilities in Unix-Style File System

在线阅读下载全文

作  者:韩伟[1,2,3] 贺也平[1] 

机构地区:[1]基础软件国家工程中心(中国科学院软件研究所),北京100190 [2]中国科学院研究生院,北京100049 [3]石家庄铁道大学信息科学与技术学院,石家庄050043

出  处:《计算机研究与发展》2011年第8期1430-1437,共8页Journal of Computer Research and Development

基  金:国家自然科学基金项目(90818012);"核高基"国家科技重大专项(2010ZX01036-001-002);中国科学院知识创新工程重要方向项目(KGCX2-YW-125)

摘  要:文件系统中的TOCTTOU缺陷是类Unix操作系统面临的一个严重安全问题,现有的静态检测方法具有很高的误报率.原因有2点:一是对导致TOCTTOU缺陷的函数对缺乏精确定义和分析;二是分析过程过度抽象,忽略了很多重要的程序信息.因此,首先对TOCTTOU缺陷进行了分类,并系统分析了C标准库中可以导致TOCTTOU缺陷的函数对.在此基础上,提出了一种TOCTTOU缺陷的静态分析方法,利用有限状态安全属性刻画TOCTTOU缺陷,分析精度达到了过程内路径敏感、过程间流敏感.实验结果表明,该方法能够有效检测C程序中的TOCTTOU缺陷,相比现有方法,有效降低了误报率.TOCTTOU is a serious threat to Unix-style file systems. All the existing static detection methods have high false positive rate. There are two reasons., firstly, the function pairs which may cause TOCTTOU vulnerabilities are not defined and enumerated accurately; and secondly, the methods make an o^eer-approximation of the program and omit a lot of useful information. In this paper, we first systematically examine the TOCTTOU pairs in the standard C library. On this basis, a static analysis method is presented to detect the TOCTTOU vulnerabilities. Vulnerability is expressed as a finite safety state property. At each program point, a value is associated to a set of states. To make the analysis more precise, the algorithm is inter-procedurally flow sensitive and intra- procedurally path sensitive. To achieve scalability, the safety state property of each procedural is analyzed independently and the inter-procedurally analysis is summary based. The experimental results show that this method can effectively find TOCTTOU vulnerabilities in C programs. In comparison with other static methods, this method can effectively reduce false positive rate.

关 键 词:TOCTTOU缺陷 文件竞争条件 静态分析 流敏感分析 路径敏感分析 

分 类 号:TP311[自动化与计算机技术—计算机软件与理论]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象