Authors:
Sangharatna Godboley
1
and
Arpita Dutta
2
Affiliations:
1
Department of CSE, NIT Warangal, Telangana, India
;
2
Department of CSE, IIT Kharagpur, West Bengal, India
Keyword(s):
MC/DC, Test Cases, Software Testing, Static MC/DC Analysis, Dynamic MC/DC Analysis.
Abstract:
RTCA/DO-178B&C standards mandate Modified Condition / Decision Coverage (MC/DC) criterion for level-A category software. In critical safety system applications such as Aircraft or Metro Rail controller systems, etc., testing engineers have to produce the MC/DC report. There are several MC/DC analyzers, which are either automated or partially-automated available. Some of the existing analyzers do not consider the dependencies of Predicates/Decisions on each other. These analyzers process each predicate individually based on MC/DC criterion. They use test cases to identify the total number of atomic conditions present in a decision which influence the output of whole decision. In this paper, we overcome the limitations of some of the existing techniques. We propose an approach, which execute the whole program along with unit test cases at run time to compute MC/DC score. This dynamic mechanism solves the dependency relation between the variables appearing at different predicates and th
eir branch statements in a single run. We have developed Dynamic COverage PErcentage CAlculator (Dy-COPECA) using C and Java language to process C-programs. We have improved the MC/DC by 42.88% through dynamic MC/DC analysis as compared to static analysis for the example C-program.
(More)