Fun and Practice for exercising your ARM(64)
summary
Presents a practical fuzzer that combines ARM64 hardware trace with Angr-based hybrid symbolic execution to reduce generated inputs and hit more driver code paths. Describes a prototype harness that gathers instruction traces on real ARM64 devices, injects trace feedback into Angr to prune constraints, and drives a mutational engine. Claims significant speedup in tracing and higher coverage versus blind fuzzing and demonstrates several driver bugs found during testing.
AI-written summary · about
related talks
54mMy Fuzzy Driver
28mkAFL: Hardware-Assisted Feedback Fuzzing for OS Kernels
32mFuzzing Zephyr Apps - Struggles of Dynamic Analysis on Embedded Applications
50mWhat the Fuzz
34m1-Click-Fuzz：Systematically Fuzzing the Windows Kernel Driver with Symbolic Execution.
01h 15mMake your programs more reliable with Fuzzing
13mFuzzing Comes to Solana
42mIdentifying Safety Weaknesses and Fault Propagation in the Linux Kernel