Compare the performance of afl-gcc instrumentation, afl-clang-fast (a.k.a llvm_mode) instrumentation, and QEMU instrumentation. Used AFL-2.56b as a fuzzer, and three binutils programs (nm, readelf, size) as targets.
echo core | sudo tee /proc/sys/kernel/core_pattern
docker build -t comparison -f Dockerfile .
docker run --rm -it comparison
./run_all.sh