PROJECT TITLE :

Faultprog: Testing the Accuracy of Binary-Level Software Fault Injection - 2018

ABSTRACT:

Off-The-Shelf (OTS) software parts are the cornerstone of contemporary systems, as well as safety-important ones. However, the dependability of OTS elements is uncertain thanks to the lack of source code, design artifacts and check cases, since only their binary code is equipped. Fault injection in elements' binary code could be a answer to understand the risks posed by buggy OTS parts. In this Project, we have a tendency to consider the problem of the accurate mutation of binary code for fault injection purposes. Fault injection emulates bugs in high-level programming constructs (assignments, expressions, perform calls, ...) by mutating their translation in binary code. But, the semantic gap between the supply code and its binary translation usually leads to inaccurate mutations. We propose Faultprog, a scientific approach for testing the accuracy of binary mutation tools. Faultprog automatically generates synthetic programs employing a stochastic grammar, and mutates both their binary code with the tool underneath check, and their supply code as reference for comparisons. Moreover, we have a tendency to present a case study on a business binary mutation tool, where Faultprog was adopted to identify code patterns and compiler optimizations that affect its mutation accuracy.