FIRM-AFL is the first high-throughput greybox fuzzer for IoT firmware.

Related tags

Deep LearningFirmAFL
Overview

FIRM-AFL

FIRM-AFL is the first high-throughput greybox fuzzer for IoT firmware. FIRM-AFL addresses two fundamental problems in IoT fuzzing. First, it addresses compatibility issues by enabling fuzzing for POSIX-compatible firmware that can be emulated in a system emulator. Second, it addresses the performance bottleneck caused by system-mode emulation with a novel technique called "augmented process emulation". By combining system-mode emulation and user-mode emulation in a novel way, augmented process emulation provides high compatibility as system-mode emulation and high throughput as user-mode emulation.

Publication

Yaowen Zheng, Ali Davanian, Heng Yin, Chengyu Song, Hongsong Zhu, Limin Sun, “FIRM-AFL: High-throughput greybox fuzzing of IoT firmware via augmented process emulation,” in USENIX Security Symposium, 2019.

Introduction

FIRM-AFL is the first high-throughput greybox fuzzer for IoT firmware. FIRM-AFL addresses two fundamental problems in IoT fuzzing. First, it addresses compatibility issues by enabling fuzzing for POSIX-compatible firmware that can be emulated in a system emulator. Second, it addresses the performance bottleneck caused by system-mode emulation with a novel technique called "augmented process emulation". By combining system-mode emulation and user-mode emulation in a novel way, augmented process emulation provides high compatibility as system-mode emulation and high throughput as user-mode emulation. The overview is show in Figure 1.

Figure 1. Overview of Augmented Process Emulation

 

We design and implement FIRM-AFL, an enhancement of AFL for fuzzing IoT firmware. We keep the workflow of AFL intact and replace the user-mode QEMU with augmented process emulation, and the rest of the components remain unchanged. The new workflow is illustrated in Figure 2.

Figure 2. Overview of FIRM-AFL

Setup

Our system has two parts: system mode and user mode. We compile them separately for now.

User mode

cd user_mode/
./configure --target-list=mipsel-linux-user,mips-linux-user,arm-linux-user --static --disable-werror
make

System mode

cd qemu_mode/DECAF_qemu_2.10/
./configure --target-list=mipsel-softmmu,mips-softmmu,arm-softmmu --disable-werror
make

Usage

  1. Download the Firmdyne repo to the root directory of FirmAFL, then setup the firmadyne according to its instructions including importing its datasheet https://cmu.app.boxcn.net/s/hnpvf1n72uccnhyfe307rc2nb9rfxmjp into database.

  2. Replace the scripts/makeImage.sh with modified one in firmadyne_modify directory.

  3. follow the guidance from firmadyne to generate the system running scripts.

Take DIR-815 router firmware as a example,

cd firmadyne
./sources/extractor/extractor.py -b dlink -sql 127.0.0.1 -np -nk "../firmware/DIR-815_FIRMWARE_1.01.ZIP" images
./scripts/getArch.sh ./images/9050.tar.gz
./scripts/makeImage.sh 9050
./scripts/inferNetwork.sh 9050
cd ..
python FirmAFL_setup.py 9050 mipsel
  1. modify the run.sh in image_9050 directory as following, in order to emulate firmware with our modified QEMU and kernel, and running on the RAM file.

For mipsel,

ARCH=mipsel
QEMU="./qemu-system-${ARCH}"
KERNEL="./vmlinux.${ARCH}_3.2.1" 
IMAGE="./image.raw"
MEM_FILE="./mem_file"
${QEMU} -m 256 -mem-prealloc -mem-path ${MEM_FILE} -M ${QEMU_MACHINE} -kernel ${KERNEL} \ 

For mipseb,

ARCH=mips
QEMU="./qemu-system-${ARCH}"
KERNEL="./vmlinux.${ARCH}_3.2.1" 
IMAGE="./image.raw"
MEM_FILE="./mem_file"
${QEMU} -m 256 -mem-prealloc -mem-path ${MEM_FILE} -M ${QEMU_MACHINE} -kernel ${KERNEL} \
  1. run the fuzzing process

after running the start.py script, FirmAFL will start the firmware emulation, and after the system initialization(120s), the fuzzing process will start. (Maybe you should use root privilege to run it.)

cd image_9050
python start.py 9050

Related Work

Our system is built on top of TriforceAFL, DECAF, AFL, and Firmadyne.

TriforceAFL: AFL/QEMU fuzzing with full-system emulation, https://github.com/nccgroup/TriforceAFL.

DECAF: "Make it work, make it right, make it fast: building a platform-neutral whole-system dynamic binary analysis platform", Andrew Henderson, Aravind Prakash, Lok Kwong Yan, Xunchao Hu, Xujiewen Wang, Rundong Zhou, and Heng Yin, to appear in the International Symposium on Software Testing and Analysis (ISSTA'14), San Jose, CA, July 2014. https://github.com/sycurelab/DECAF.

AFL: american fuzzy lop (2.52b), http://lcamtuf.coredump.cx/afl/.

Firmadyne: Daming D. Chen, Maverick Woo, David Brumley, and Manuel Egele. “Towards automated dynamic analysis for Linux-based embedded firmware,” in Network and Distributed System Security Symposium (NDSS’16), 2016. https://github.com/firmadyne.

Troubleshooting

(1) error: static declaration of ‘memfd_create’ follows non-static declaration

Please see https://blog.csdn.net/newnewman80/article/details/90175033.

(2) failed to find romfile "efi-e1000.rom" when run the "run.sh"

Use the run.sh in FirmAFL_config/9050/ instead.

(3) Fork server crashed with signal 11

Run scripts in start.py sequentially. First run "run.sh", when the testing program starts, run "python test.py", and "user.sh".

(4) For the id "12978", "16116" firmware, since these firmware have more than 1 test case, so we use different image directory name to distinguish them.

Before FirmAFL_setup, 
first, change image directory name image_12978 to image_129780, 
then modify the firmadyne/scratch/12978 to firmadyne/scratch/129780
After that, run python FirmAFL_setup.py 129780 mips
(If you want to test another case for image_12978, you can use image_129781 instead image_129780)
Context-Aware Image Matting for Simultaneous Foreground and Alpha Estimation

Context-Aware Image Matting for Simultaneous Foreground and Alpha Estimation This is the inference codes of Context-Aware Image Matting for Simultaneo

Qiqi Hou 125 Oct 22, 2022
Leveraging OpenAI's Codex to solve cornerstone problems in Music

Music-Codex Leveraging OpenAI's Codex to solve cornerstone problems in Music Please NOTE: Presented generated samples were created by OpenAI's Codex P

Alex 2 Mar 11, 2022
GAN Image Generator and Characterwise Image Recognizer with python

MODEL SUMMARY 모델의 구조는 크게 6단계로 나뉩니다. STEP 0: Input Image Predict 할 이미지를 모델에 입력합니다. STEP 1: Make Black and White Image STEP 1 은 입력받은 이미지의 글자를 흑색으로, 배경을

Juwan HAN 1 Feb 09, 2022
Breast Cancer Detection 🔬 ITI "AI_Pro" Graduation Project

BreastCancerDetection - This program is designed to predict two severity of abnormalities associated with breast cancer cells: benign and malignant. Mammograms from MIAS is preprocessed and features

6 Nov 29, 2022
A simple implementation of Kalman filter in single object tracking

kalman-filter-in-single-object-tracking A simple implementation of Kalman filter in single object tracking https://www.bilibili.com/video/BV1Qf4y1J7D4

130 Dec 26, 2022
Human Dynamics from Monocular Video with Dynamic Camera Movements

Human Dynamics from Monocular Video with Dynamic Camera Movements Ri Yu, Hwangpil Park and Jehee Lee Seoul National University ACM Transactions on Gra

215 Jan 01, 2023
Fastshap: A fast, approximate shap kernel

fastshap: A fast, approximate shap kernel fastshap was designed to be: Fast Calculating shap values can take an extremely long time. fastshap utilizes

Samuel Wilson 22 Sep 24, 2022
Deep Face Recognition in PyTorch

Face Recognition in PyTorch By Alexey Gruzdev and Vladislav Sovrasov Introduction A repository for different experimental Face Recognition models such

Alexey Gruzdev 141 Sep 11, 2022
MISSFormer: An Effective Medical Image Segmentation Transformer

MISSFormer Code for paper "MISSFormer: An Effective Medical Image Segmentation Transformer". Please read our preprint at the following link: paper_add

Fong 22 Dec 24, 2022
Repo for "Benchmarking Robustness of 3D Point Cloud Recognition against Common Corruptions" https://arxiv.org/abs/2201.12296

Benchmarking Robustness of 3D Point Cloud Recognition against Common Corruptions This repo contains the dataset and code for the paper Benchmarking Ro

Jiachen Sun 168 Dec 29, 2022
Hypercomplex Neural Networks with PyTorch

HyperNets Hypercomplex Neural Networks with PyTorch: this repository would be a container for hypercomplex neural network modules to facilitate resear

Eleonora Grassucci 21 Dec 27, 2022
Orbivator AI - To Determine which features of data (measurements) are most important for diagnosing breast cancer and find out if breast cancer occurs or not.

Orbivator_AI Breast Cancer Wisconsin (Diagnostic) GOAL To Determine which features of data (measurements) are most important for diagnosing breast can

anurag kumar singh 1 Jan 02, 2022
Source code for "Interactive All-Hex Meshing via Cuboid Decomposition [SIGGRAPH Asia 2021]".

Interactive All-Hex Meshing via Cuboid Decomposition Video demonstration This repository contains an interactive software to the PolyCube-based hex-me

Lingxiao Li 131 Dec 05, 2022
Message Passing on Cell Complexes

CW Networks This repository contains the code used for the papers Weisfeiler and Lehman Go Cellular: CW Networks (Under review) and Weisfeiler and Leh

Twitter Research 108 Jan 05, 2023
A flexible submap-based framework towards spatio-temporally consistent volumetric mapping and scene understanding.

Panoptic Mapping This package contains panoptic_mapping, a general framework for semantic volumetric mapping. We provide, among other, a submap-based

ETHZ ASL 194 Dec 20, 2022
NumPy로 구현한 딥러닝 라이브러리입니다. (자동 미분 지원)

Deep Learning Library only using NumPy 본 레포지토리는 NumPy 만으로 구현한 딥러닝 라이브러리입니다. 자동 미분이 구현되어 있습니다. 자동 미분 자동 미분은 미분을 자동으로 계산해주는 기능입니다. 아래 코드는 자동 미분을 활용해 역전파

조준희 17 Aug 16, 2022
quantize aware training package for NCNN on pytorch

ncnnqat ncnnqat is a quantize aware training package for NCNN on pytorch. Table of Contents ncnnqat Table of Contents Installation Usage Code Examples

62 Nov 23, 2022
Learning Visual Words for Weakly-Supervised Semantic Segmentation

[IJCAI 2021] Learning Visual Words for Weakly-Supervised Semantic Segmentation Implementation of IJCAI 2021 paper Learning Visual Words for Weakly-Sup

Lixiang Ru 24 Oct 05, 2022
we propose a novel deep network, named feature aggregation and refinement network (FARNet), for the automatic detection of anatomical landmarks.

Feature Aggregation and Refinement Network for 2D Anatomical Landmark Detection Overview Localization of anatomical landmarks is essential for clinica

aoyueyuan 0 Aug 28, 2022
Complete* list of autonomous driving related datasets

AD Datasets Complete* and curated list of autonomous driving related datasets Contributing Contributions are very welcome! To add or update a dataset:

Daniel Bogdoll 13 Dec 19, 2022