Non-intrusive MC/DC measurement based on traces

Faustin Ahishakiye; Svetlana Jaksic; Felix Dino Lange; Malte Schmitz; Volker Stolz; Daniel Thoma

doi:10.1109/TASE.2019.00-15

Non-intrusive MC/DC measurement based on traces

Faustin Ahishakiye, Svetlana Jaksic, Felix Dino Lange, Malte Schmitz, Volker Stolz, Daniel Thoma

Institute of Software Engineering and Programming Languages

Western Norway University of Applied Sciences

Abstract

We present a novel, non-intrusive approach to MC/DC coverage measurement using modern processor-based tracing facilities. Our approach does not require recompilation or instrumentation of the software under test. Instead, we use the Intel Processor Trace (Intel PT) facility present on modern Intel CPUs. Our tooling consists of the following parts: a frontend that detects so-called decisions (Boolean expressions) that are used in conditionals in C source code, a mapping from conditional jumps in the object code back to those decisions, and an analysis that computes satisfaction of the MC/DC coverage relation on those decisions from an execution trace. This analysis takes as input a stream of instruction addresses decoded from Intel PT trace data, which was recorded while running the software under test. We describe our architecture and discuss limitations and future work.

Original language	English
Title of host publication	2019 International Symposium on Theoretical Aspects of Software Engineering (TASE)
Number of pages	7
Publisher	IEEE
Publication date	07.2019
Pages	86-92
Article number	8914094
ISBN (Print)	978-1-7281-3343-0
ISBN (Electronic)	978-1-7281-3342-3
DOIs	https://doi.org/10.1109/TASE.2019.00-15
Publication status	Published - 07.2019
Event	13th International Symposium on Theoretical Aspects of Software Engineering - Guilin, China Duration: 29.07.2019 → 31.07.2019 Conference number: 155650

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/TASE.2019.00-15

Cite this

@inproceedings{93b71e48d2cb4366b115a622bc3393e2,

title = "Non-intrusive MC/DC measurement based on traces",

abstract = "We present a novel, non-intrusive approach to MC/DC coverage measurement using modern processor-based tracing facilities. Our approach does not require recompilation or instrumentation of the software under test. Instead, we use the Intel Processor Trace (Intel PT) facility present on modern Intel CPUs. Our tooling consists of the following parts: a frontend that detects so-called decisions (Boolean expressions) that are used in conditionals in C source code, a mapping from conditional jumps in the object code back to those decisions, and an analysis that computes satisfaction of the MC/DC coverage relation on those decisions from an execution trace. This analysis takes as input a stream of instruction addresses decoded from Intel PT trace data, which was recorded while running the software under test. We describe our architecture and discuss limitations and future work.",

author = "Faustin Ahishakiye and Svetlana Jaksic and Lange, {Felix Dino} and Malte Schmitz and Volker Stolz and Daniel Thoma",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; 13th International Symposium on Theoretical Aspects of Software Engineering, TASE 2019 ; Conference date: 29-07-2019 Through 31-07-2019",

year = "2019",

month = jul,

doi = "10.1109/TASE.2019.00-15",

language = "English",

isbn = "978-1-7281-3343-0",

pages = "86--92",

booktitle = "2019 International Symposium on Theoretical Aspects of Software Engineering (TASE)",

publisher = "IEEE",

}

Ahishakiye, F, Jaksic, S, Lange, FD, Schmitz, M, Stolz, V & Thoma, D 2019, Non-intrusive MC/DC measurement based on traces. in 2019 International Symposium on Theoretical Aspects of Software Engineering (TASE)., 8914094, IEEE, pp. 86-92, 13th International Symposium on Theoretical Aspects of Software Engineering, Guilin, China, 29.07.19. https://doi.org/10.1109/TASE.2019.00-15

TY - GEN

T1 - Non-intrusive MC/DC measurement based on traces

AU - Ahishakiye, Faustin

AU - Jaksic, Svetlana

AU - Lange, Felix Dino

AU - Schmitz, Malte

AU - Stolz, Volker

AU - Thoma, Daniel

N1 - Conference code: 155650

PY - 2019/7

Y1 - 2019/7

N2 - We present a novel, non-intrusive approach to MC/DC coverage measurement using modern processor-based tracing facilities. Our approach does not require recompilation or instrumentation of the software under test. Instead, we use the Intel Processor Trace (Intel PT) facility present on modern Intel CPUs. Our tooling consists of the following parts: a frontend that detects so-called decisions (Boolean expressions) that are used in conditionals in C source code, a mapping from conditional jumps in the object code back to those decisions, and an analysis that computes satisfaction of the MC/DC coverage relation on those decisions from an execution trace. This analysis takes as input a stream of instruction addresses decoded from Intel PT trace data, which was recorded while running the software under test. We describe our architecture and discuss limitations and future work.

AB - We present a novel, non-intrusive approach to MC/DC coverage measurement using modern processor-based tracing facilities. Our approach does not require recompilation or instrumentation of the software under test. Instead, we use the Intel Processor Trace (Intel PT) facility present on modern Intel CPUs. Our tooling consists of the following parts: a frontend that detects so-called decisions (Boolean expressions) that are used in conditionals in C source code, a mapping from conditional jumps in the object code back to those decisions, and an analysis that computes satisfaction of the MC/DC coverage relation on those decisions from an execution trace. This analysis takes as input a stream of instruction addresses decoded from Intel PT trace data, which was recorded while running the software under test. We describe our architecture and discuss limitations and future work.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85076978483&origin=inward&txGid=ce4d9e3ff546cb79f9186dad9a1cab19

U2 - 10.1109/TASE.2019.00-15

DO - 10.1109/TASE.2019.00-15

M3 - Conference contribution

SN - 978-1-7281-3343-0

SP - 86

EP - 92

BT - 2019 International Symposium on Theoretical Aspects of Software Engineering (TASE)

PB - IEEE

T2 - 13th International Symposium on Theoretical Aspects of Software Engineering

Y2 - 29 July 2019 through 31 July 2019

ER -

Non-intrusive MC/DC measurement based on traces

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this