Linux Testing made better with DATA
summary
Uses ctags to index kernel functions and a G code tool to collect per-test code coverage, stores source and test databases, and runs a Jenkins job that parses a patch to emit a dynamic test suite. Claims coverage-based reduction up to 99% and reports initial runs with ~93.5% reduction and one case where ~1,000 tests drop to 129 while preserving fault detection; also flags low-coverage files to prioritise new or improved tests.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:27Agenda
- 5:20Overview
- 5:37Source Database Creation
- 9:30How To Know What Test Cases To Be Improved
- 11:42Disclaimer Chart
- 13:36Is There any Way We Can Measure Path Coverage Rather than Line Coverage
related talks
15m"Linux Testing made better with DATA"
37mKernel CI – How Far Can It Go?
27mFaster Linux Kernel Testing with Linaro's Open-Source Tools and LKFT Automation
33mKernel CI – How Far Can It Go?
41mThe magical fantasy land of Linux kernel testing
19mTesting Your Tests With Code Coverage
50mTips for Writing Good Tests for Linux
19mChecking your work: Linux kernel testing and CI