SearchGEM5: Towards Reliable gem5 with Search Based Software Testing and Large Language Models

Aidan Dakhama, Karine Even-Mendoza, William B. Langdon, Hector D. Menendez, Justyna Petke

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

49 Downloads (Pure)

Abstract

We introduce a novel automated testing technique that combines LLM and search-based fuzzing. We use ChatGPT to parameterise C programs. We compile the resultant code snippets, and feed compilable ones to SearchGEM5 — our extension to AFL++ fuzzer with customised new mutation operators. We run thus created 4005 binaries through our system under test, gem5, increasing its existing test coverage by more than 1000 lines. We discover 244 instances where gem5 simulation of the binary differs from the binary’s expected behaviour.
Original languageEnglish
Title of host publication15th Symposium on Search Based Software Engineering (SSBSE)
Subtitle of host publicationLecture Notes in Computer Science
Place of PublicationUnited States
PublisherSpringer
Number of pages6
Volume15
Publication statusPublished - 8 Dec 2023

Keywords

  • AI
  • LLM
  • SBSE
  • SBFT
  • genetic improvement of tests
  • gem5

Fingerprint

Dive into the research topics of 'SearchGEM5: Towards Reliable gem5 with Search Based Software Testing and Large Language Models'. Together they form a unique fingerprint.

Cite this