Static analysis method and system based on symbolic function abstracts

Abstract

The invention discloses a static analysis method based on symbolic function abstracts. The method includes the steps that an RSTVL model is utilized to describe the storage state of variables of nodes of a control flow diagram of a current function; when the current node of the function is determined as a non-final node and functions which can be called exist in the current node, the function abstracts of the functions called by the current function are instantiated, and variables affected by call of the functions and in the current function are updated; when the current node of the function is determined as a final node and the current function has a function return value, a symbol expression of the function return value is acquired, the variables of which the storage state changes are found out, and symbol expressions of the variables of which the storage state changes are acquired; the symbol expressions of the function return value and the symbol expressions of external variables in the variables of which the storage state changes are added to the function abstracts of the current function. The invention further discloses a static analysis system based on the symbolic function abstracts. By the adoption of the technical scheme, static analysis accuracy and efficiency can be effectively improved.

Description

technical field [0001] The invention relates to a static analysis technology, in particular to a static analysis method and system based on a symbolized function summary. Background technique [0002] By testing the software program, the shortcomings and defects of the program itself can be found. Among them, software testing methods include: static analysis and dynamic analysis methods. [0003] The static analysis method is also called static testing, which refers to not actually running the software under test, but scanning the source program to find out the abnormal structure, abnormal control flow and abnormal data flow of the program. It can be seen that static testing of programs is an important part of ensuring software quality. At the same time, the static analysis method is favored due to its advantages of low cost, simple implementation, and coverage of all paths. [0004] Analysis accuracy and analysis efficiency are two important indicators for evaluating sta...

AI Technical Summary

Problems solved by technology

The current inter-procedural static analysis methods include: process inlining, recording function call string, function summary, etc. Among them, the static analysis method based on function summary can realize the sensitive analysis of the program context, but because the program usually contains many Complex data type variables such as level pointer variables and compound variables lead to insufficient analysis accuracy and analysis efficiency

Images