Method for picking the smallest subset of code that covers all source code structures

Abstract

The present invention provides a method for selecting the smallest code subset covering all source code structures, which includes: a. classifying the source code of the software to be analyzed according to high-level language, assembly language and library functions to obtain a high-level language code set, A collection of assembly codes and a collection of library functions; b. analyzing the grammatical structure of each function in the collection of high-level language codes and the structure of additional codes that may be generated during compilation, and selecting under the condition that each structure is covered at least once extracting a typical structural code subset that minimizes the workload of source code object code comparative analysis; and c. merging the typical structural code subset, the assembly code set and the library function set to obtain the software to be analyzed The minimal code subset of .

Description

technical field [0001] The invention mainly relates to the field of aviation airborne software verification, in particular to a method for selecting the smallest code subset covering all source code structures. Background technique [0002] In the design of civil aviation airborne software, A-level software is the software with the highest safety level, and its safety and reliability must be ensured. [0003] Both FAA and CAAC recognize DO-178C, "Software Considerations in Airborne Systems and Equipment Qualification Certification" as the standard for airworthiness certification of airborne software. Verification Goal". [0004] At present, the functions of airborne software are becoming more and more complex, and the scale is also increasing. For A-level software, it is necessary to identify additional codes introduced by compilers, linkers or other tools. In existing solutions, either based on source code-level structural coverage testing and supplementing source code-ob...

AI Technical Summary

Problems solved by technology

Because the same program structure exists in the software to be analyzed or the software to be analyzed does not involve the combination of structures under all coding rules, the use of these two methods will lead to manual identification of source code and object code differences due to redundant grammatical structures. volume increase

[0005] Simultaneously in the prior art, as CN104360949A discloses a kind of conforming to DO-178B / C A level software target code coverage analysis method, but it needs first to create the encoding rule table by identifying the source code target code correspondence relation and source code target code difference And extra mapping table, carry out target code coverage analysis again, it does not provide specific and operable typical source code structure definition and selection algorithm; CN104461882A discloses a kind of model verification method that meets DO-178B / C A level, but it is in the model In the field of verification, it involves a method for selecting a subset of verification programs from a generated verification program set. This verification program is a test operation step that is equivalent to the concept of a software test case, and is not a verified software program; CN104731587A discloses a unit test data generation Method, by analyzing the structure of the source code to determine the type and structure of the test data, and generate test data, the source code structure analysis does not involve the selection of source code subsets using the structure analysis results

[0006] In the existing method, as the scale of software increases, the workload and cost of comparative analysis of source code and object code will increase in order to "meet the goal of verifying additional code that cannot be traced back to the source code".

Images