AntiFuzz: Impeding Fuzzing Audits of Binary Executables

Related tags

Deep Learningantifuzz
Overview

AntiFuzz: Impeding Fuzzing Audits of Binary Executables

Get the paper here: https://www.usenix.org/system/files/sec19-guler.pdf

Usage:

The python script antifuzz_generate.py generates a "antifuzz.h" file that you need to include in your C project (see chapter below). The script takes multiple arguments to define which features you want to activate.

To disable all features, supply:

  --disable-all

To break assumption (A), i.e. to break coverage-guided fuzzing, use:

  --enable-anti-coverage

You can specify how many random BBs and random constrain functions you want to have by supplying "--anti-coverage [num]" (default: 10000).

To break assumption (B), i.e. to prevent fuzzers from detecting crashes, use:

  --signal --crash-action exit

To break assumption (C), i.e. to decrease the performance of the application when being fuzzed, use:

  --enable-sleep --signal

Additionaly, you can supply "--sleep [ms]" to set the length of the sleep in milliseconds (default: 750). You can also replace the crash behavior by supplying "--crash-action timeout" to replace every crash with a timeout.

To break assumption (D), i.e. to boggle down symbolic execution engines, use:

  --hash-cmp --enable-encrypt-decrypt

To enable all features, use:

  --enable-anti-coverage --signal --crash-action exit --enable-sleep --signal --hash-cmp --enable-encrypt-decrypt

Demo

To test it out, we supplied a demo application called antifuzz_test.c that just checks for "crsh" with single byte comparisons, and crashes if that's the case. It configures itself to fit the generated antifuzz header file, i.e. when hash comparisons are demanded via antifuzz_generate.py, antifuzz_test will compare the hashes instead of the plain constants.

First, generate the antifuzz.h file:

python antifuzz_generate.py --enable-anti-coverage --signal --crash-action exit --enable-sleep --signal --hash-cmp --enable-encrypt-decrypt

Next, compile the demo application with afl-gcc after installing AFL 2.52b (note that this may take minutes (!) depending on the number of random BBs added):

afl-gcc antifuzz_test.c -o antifuzz_test 

Run it in AFL to test it out:

mkdir inp; echo 1234 > inp/a.txt; afl-fuzz -i inp/ -o /dev/shm/out -- ./antifuzz_test @@

If you enabled all options, AFL may take a long time to start because the application is slowed down (to break assumption (C))

Protecting Applications

To include it in your own C project, follow these instructions (depending on your use-case and application, you might want to skip some of them):

1.

Add

#include "antifuzz.h"

to the header.

2.

Jump to the line that opens the (main) input file, the one that an attacker might target as an attack vector, and call

antifuzz_init("file_name_here", FLAG_ALL); 

This initializes AntiFuzz, checks if overwriting signals is possible, checks if the application is ptrace'd, puts the input through encryption and decryption, jumps through random BBs, etc.

3.

Find all lines and blocks of code that deal with malformed input files or introduce those yourself. It's often the case that these lines already exist to print some kind of error or warning message (e.g. "this is not a valid ... file"). Add a call to

antifuzz_onerror()

everywhere you deem appropriate.

4.

Find comparisons to constants (e.g. magic bytes) that you think are important for this file format, and change the comparison to hash comparisons. Add your constant to antifuzz_constants.tpl.h like this:

char *antifuzzELF = "ELF";

Our generator script will automatically change these lines to their respective SHA512 hashes when generating the final header file, you do not have to do this manually. Now change the lines from (as an example):

if(strcmp(header, "ELF") == 0)

to

if(antifuzz_str_equal(header, antifuzzELF))

See antifuzz.tpl.h for more comparison functions.

5.

If you have more data that you want to protect from symbolic execution, use:

antifuzz_encrypt_decrypt_buf(char *ptr, size_t fileSize) 
Owner
Chair for Sys­tems Se­cu­ri­ty
Chair for Sys­tems Se­cu­ri­ty
A CNN implementation using only numpy. Supports multidimensional images, stride, etc.

A CNN implementation using only numpy. Supports multidimensional images, stride, etc. Speed up due to heavy use of slicing and mathematical simplification..

2 Nov 30, 2021
Transformer Tracking (CVPR2021)

TransT - Transformer Tracking [CVPR2021] Official implementation of the TransT (CVPR2021) , including training code and trained models. We are revisin

chenxin 465 Jan 06, 2023
ConvMixer unofficial implementation

ConvMixer ConvMixer 非官方实现 pytorch 版本已经实现。 nets 是重构版本 ,test 是官方代码 感兴趣小伙伴可以对照看一下。 keras 已经实现 tf2.x 中 是tensorflow 2 版本 gelu 激活函数要求 tf=2.4 否则使用入下代码代替gelu

Jian Tengfei 8 Jul 11, 2022
Denoising Diffusion Probabilistic Models

Denoising Diffusion Probabilistic Models Jonathan Ho, Ajay Jain, Pieter Abbeel Paper: https://arxiv.org/abs/2006.11239 Website: https://hojonathanho.g

Jonathan Ho 1.5k Jan 08, 2023
Robust Self-augmentation for NER with Meta-reweighting

Robust Self-augmentation for NER with Meta-reweighting

Lam chi 17 Nov 22, 2022
Expert Finding in Legal Community Question Answering

Expert Finding in Legal Community Question Answering Arian Askari, Suzan Verberne, and Gabriella Pasi. Expert Finding in Legal Community Question Answ

Arian Askari 3 Oct 31, 2022
ZeroGen: Efficient Zero-shot Learning via Dataset Generation

ZEROGEN This repository contains the code for our paper “ZeroGen: Efficient Zero

Jiacheng Ye 31 Dec 30, 2022
NeuralCompression is a Python repository dedicated to research of neural networks that compress data

NeuralCompression is a Python repository dedicated to research of neural networks that compress data. The repository includes tools such as JAX-based entropy coders, image compression models, video c

Facebook Research 297 Jan 06, 2023
Caffe: a fast open framework for deep learning.

Caffe Caffe is a deep learning framework made with expression, speed, and modularity in mind. It is developed by Berkeley AI Research (BAIR)/The Berke

Berkeley Vision and Learning Center 33k Dec 28, 2022
High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.

TL;DR Ignite is a high-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently. Click on the image to

4.2k Jan 01, 2023
FOSS Digital Asset Distribution Platform built on Frappe.

Digistore FOSS Digital Assets Marketplace. Distribute digital assets, like a pro. Video Demo Here Features Create, attach and list digital assets (PDF

Mohammad Hussain Nagaria 30 Dec 08, 2022
Collective Multi-type Entity Alignment Between Knowledge Graphs (WWW'20)

CG-MuAlign A reference implementation for "Collective Multi-type Entity Alignment Between Knowledge Graphs", published in WWW 2020. If you find our pa

Bran Zhu 28 Dec 11, 2022
code for Grapadora research paper experimentation

Road feature embedding selection method Code for research paper experimentation Abstract Traffic forecasting models rely on data that needs to be sens

Eric López Manibardo 0 May 26, 2022
Taichi Course Homework Template

太极图形课S1-标题部分 这个作业未来或将是你的开源项目,标题的内容可以来自作业中的核心关键词,让读者一眼看出你所完成的工作/做出的好玩demo 如果暂时未想好,起名时可以参考“太极图形课S1-xxx作业” 如下是作业(项目)展开说明的方法,可以帮大家理清思路,并且也对读者非常友好,请小伙伴们多多参

TaichiCourse 30 Nov 19, 2022
Multi-label classification of retinal disorders

Multi-label classification of retinal disorders This is a deep learning course project. The goal is to develop a solution, using computer vision techn

Sundeep Bhimireddy 1 Jan 29, 2022
LaneDet is an open source lane detection toolbox based on PyTorch that aims to pull together a wide variety of state-of-the-art lane detection models

LaneDet is an open source lane detection toolbox based on PyTorch that aims to pull together a wide variety of state-of-the-art lane detection models. Developers can reproduce these SOTA methods and

TuZheng 405 Jan 04, 2023
Detector for Log4Shell exploitation attempts

log4shell-detector Detector for Log4Shell exploitation attempts Idea The problem with the log4j CVE-2021-44228 exploitation is that the string can be

Florian Roth 729 Dec 25, 2022
🔎 Monitor deep learning model training and hardware usage from your mobile phone 📱

Monitor deep learning model training and hardware usage from mobile. 🔥 Features Monitor running experiments from mobile phone (or laptop) Monitor har

labml.ai 1.2k Dec 25, 2022
The code of NeurIPS 2021 paper "Scalable Rule-Based Representation Learning for Interpretable Classification".

Rule-based Representation Learner This is a PyTorch implementation of Rule-based Representation Learner (RRL) as described in NeurIPS 2021 paper: Scal

Zhuo Wang 53 Dec 17, 2022
Machine Learning Models were applied to predict the mass of the brain based on gender, age ranges, and head size.

Brain Weight in Humans Variations of head sizes and brain weights in humans Kaggle dataset obtained from this link by Anubhab Swain. Image obtained fr

Anne Livia 1 Feb 02, 2022