Exploring and Assessing the Trivial Compiler Equivalence

Mutation testing is the state-of-the-art technique to assess the fault-detection capability of a test suite. However, its adoption in industry is deterred by few of its inherent limitations including the equivalent mutants. Since the equivalent mutants are functionally similar to the original progra...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:University of Sindh journal of information and communication technology (Online) 2022-05, Vol.6 (1), p.22-24
Hauptverfasser: Habibullah Nangraj, Gulsher Laghari
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Mutation testing is the state-of-the-art technique to assess the fault-detection capability of a test suite. However, its adoption in industry is deterred by few of its inherent limitations including the equivalent mutants. Since the equivalent mutants are functionally similar to the original program, the test suite cannot kill them, hence they produce false alarms for the developers and reduce the mutation score. Although to automatically verify whether the mutant is equivalent to the original program is undecidable, yet there exist heuristics such as trivial compiler equivalence to automatically eliminate sufficient equivalent mutants. In this paper, we explore the use of compiler optimizations at assembly level code to detect equivalent mutants and find that it can indeed detect equivalent mutants
ISSN:2521-5582
2521-5582