Skip to content

Latest commit

 

History

History
104 lines (75 loc) · 2.44 KB

INSTALL.md

File metadata and controls

104 lines (75 loc) · 2.44 KB
Raw

Installation

Requirements

  • Linux with Python 3.7+
  • Z3 solver

Clone repository

To start building Fuzzle, first clone the git repo:

$ git clone https://github.com/SoftSec-KAIST/Fuzzle
$ cd Fuzzle

Installing Dependencies

To install dependencies for generating benchmarks, run the following from Fuzzle directory:

$ python3 -m pip install -r ./maze-gen/requirements.txt

Additionally, you will need to have Z3 solver installed for handling .smt2 files from CVEs. Below are the instructions for installing Z3 solver using pySMT's install command. Note that pySMT should have been installed in the previous step as one of the dependencies for Fuzzle.

$ pysmt-install --z3

To check that the solver was installed correctly, you can use the following command:

$ pysmt-install --check

Building Docker Images

To run the fuzzers on the generated benchmark, you must first build the corresponding docker images:

$ ./scripts/build_all_dockers.sh

Note that this may take a few hours depending on your machine.

Basic Usage Example

Once the installation is complete, you can quickly check that everything is working correctly by running test.sh.

$ ./test/test.sh

This will first generate a benchmark program, run AFL, save results, and generate coverage visualization. The output will be automatically saved in test_output directory under test directory.

Note that this script will take approximately 16 minutes in total.

After the script finishes running, you should have something similar to the below printed out to the screen:

Measure:        Coverage (%)
################################
Tool            Algorithm
                Wilsons
--------------------------------
afl             40.0
--------------------------------

Measure:        Bugs (%)
################################
Tool            Algorithm
                Wilsons
--------------------------------
afl             0
--------------------------------

Measure:        TTE (h)
################################
Tool            Algorithm
                Wilsons
--------------------------------
afl             -
--------------------------------

The test_output directory should also contain visual.png which is the visual aid generated from the run.

Additional Notes

  • The installation and run scripts for Fuzzle were tested on Ubuntu 20.04.
  • Building docker images and running fuzzers using run_tools.py require superuser access.