- download, compile and install afl => https://github.com/mirrorer/afl
- download, compile and install dyninst => https://github.com/dyninst/dyninst
- download, compile and install afl-dyninst => https://github.com/vanhauser-thc/afl-dyninst
- download, compile and install dynamorio => https://github.com/DynamoRIO/dynamorio
- make a symlink to the afl folder here named "afl" , e.g. "ln -s ../afl afl"
- export DYNAMORIO_HOME=/path/to/dynamorio/build directory
- make install
How to run
afl-dyninst.sh -i program -o program_instrumented -D It is a good idea to add -e and -E with well selected function addresses to make the fuzzing faster NOTE: you can skip this step and use -forkserver option in the next step. But this is slower at the moment!
afl-fuzz-dynamorio.sh [normal afl-fuzz options] That's it! If you fuzzing does not run, afl-fuzz might need more memory, set AFL_MEM to a high value, e.g. 700 for 700MB
When to use it
when normal afl-dyninst is crashing the binary and qemu mode -Q is not an option. Dynamorio is x10 slower than Qemu, 25x slower than dyninst - however 10x faster than Pintool, and works additionally on ARM and AARCH64. In memory fuzzing (function fuzzing) is a much faster option and implemented in a future release.