Preview: Verification (Correctness & Debugging)
summary
Five systems present concrete results: Pilot applies Python metadata, bytecode edits and dynamic profiling to patch online services without restarts; a fuzzing paper reduces instrumentation overhead to about one extra instruction per site using static analysis and vectorized coverage processing; a TCP-focused fuzzer uses dependency-driven seed generation, state-transition feedback and differential checking to find semantic bugs; IMD's static analysis finds 120 previously unknown memory leaks in Linux early-exit paths; Argus uses annotated causal traces and beam search to pinpoint 12 spinning‑wheel performance bugs, 10 previously unknown.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:38Bugs
- 1:34Software correctness
- 3:12Static checking
- 4:14Formal verification
- 4:59Model checking
- 7:17Testing
- 9:13Logging
- 9:47Dynamic Tracking
- 12:30Riff
- 13:25TCP
- 14:36Memory Leak Detection
- 15:52Debugging Performance Issues
related talks
52mStructure aware fuzzing
53mThree Colours of Fuzzing: Reflections and Open Challenges
26mExposing Hidden Exploitable Behaviors In Programming Languages Using Differential Fuzzing
38mTesting with two failure seeking missiles: fuzzing and property based testing
40mCatching More Bugs with Fewer False Alarms
42mSurprise yourself with Property Based Testing
01h 07mSufficiently Advanced Testing
45mFinding Vulnerabilities through Static Analysis and Scripting