Coder Social home page Coder Social logo

armx's Introduction

ARM-X Firmware Emulation Framework

by Saumil Shah @therealsaumil

March 2020

ARMX

The ARM-X Firmware Emulation Framework is a collection of scripts, kernels and filesystems to be used with QEMU to emulate ARM/Linux IoT devices. ARM-X is aimed to facilitate IoT research by virtualising as much of the physical device as possible. It is the closest we can get to an actual IoT VM.

Devices successfully emulated with ARM-X so far:

  • D-Link DIR-880L Wi-Fi Router
  • Netgear Nighthawk R6250 Wi-Fi Router
  • Netgear Nighthawk R6400 Wi-Fi Router
  • Trivision NC227WF Wireless IP Camera
  • Cisco RV130 Wi-Fi Router

Precursors of ARM-X have been used in Saumil Shah's popular ARM IoT Exploit Laboratory training classes where students have found four 0-day vulnerabilities in various ARM/Linux IoT devices.

ARM-X Architecture and Operations

ARM-X is a collection of scripts, kernels and filesystems residing in the /armx directory. It uses qemu-system-arm to boot up a virtual ARM/Linux environment. The /armx directory is exported over NFS to also make the contents available within the QEMU guest.

The host system running qemu-system-arm is assigned the IP address 192.168.100.1 and the QEMU guest is assigned 192.168.100.2 via tap0 interface.

Architecture

The /armx directory is organised as follows:

Directory Structure

  • devices: This file contains device definitions, one per line.
  • qemuopts: Abstracted QEMU options definitions for various types of QEMU Machines.
  • run/: This folder contains scripts necessary to parse the device configuration, preload nvram contents and eventually invoke the userland processes of the device being emulated.
  • run/launcher: The main script. launcher parses the devices file and displays a menu of registered devices. Selecting one of the devices will in turn invoke qemu-system-arm with the pre-defined QEMU options, corresponding Linux kernel and extracted root file system registered with the device.
  • template/: Sample configuration and layout for a new device. Make a copy of the template when beginning to emulate a new IoT device.

Each emulated device contains the following files/directories:

  • config: Contains the device's name and description, ASLR settings, location of its root file system and commands to issue after the kernel has booted up and transferred control to the userland.
  • nvram.ini: Contents of the device's non volatile memory, used for storing configuration settings. Contents of nvram.ini are preloaded into the emulated nvram before invoking the userland init scripts.
  • kernel/: Contains a Linux kernel compiled (mostly via Buildroot) to closely match the properties of the emulated device such as kernel version, CPU support, VM_SPLIT, supported peripherals, etc.
  • rootfs/: Populated with the uncompressed file system extracted from the device's firmware.
  • run-init: Script invoked by the launcher.

The diagram below describes each stage of ARM-X:

ARM-X Operations

  1. Invoke /armx/run/launcher. This will display a menu as shown below. In this example, we select the Trivision TRI227WF Wireless IP Camera.

ARM-X Launcher

  1. Selecting one of the devices will launch it under QEMU. The kernel which is included in the kernel/ directory of the Trivision IP Camera's device configuration, is booted in qemu-system-arm and uses a pre-built Buildroot filesystem, which is referred to as hostfs.ext2. Host and guest IP addresses are assigned to 192.168.100.1 and 192.168.100.2 respectively.

ARM-X Kernel Boot Up

  1. hostfs.ext2 contains several scripts and tools useful for running and dynamic analysis of the emulated device. The init scripts in hostfs.ext2 mount the /armx directory over NFS. Thus, the contents of /armx are shared by both the host and the QEMU guest.

  2. To kick off the rest of the device startup, connect to the QEMU guest using SSH ssh [email protected]. This brings up a menu as shown below:

ARM-X Trivision Init

  1. Selecting the option to launch the userland init scripts of the device results in run-init being invoked from the corresponding device configuration directory within /armx. First, the contents of nvram.ini are loaded into the kernel's emulated nvram driver. Next, a chroot jail is created using the rootfs of the device. Lastly, the registered initialisation commands are invoked in the newly chrooted rootfs, bringing up the device's services and init scripts.

ARM-X Trivision Started

  1. Once the device has fully "booted up" in ARM-X, it is available for testing and analysis. The image below shows the administration interface of the IP Camera loaded in a browser:

ARM-X Admin Interface

Creating your own emulated IoT Device

Before you begin to emulate an IoT device, you will need the following:

  • Detailed analysis of the IoT device
  • CPU (ARMv5/ARMv6/ARMv7)
  • Linux Kernel version
  • Extracted mtdblocks (rootfs)
  • Contents of nvram
  • Generate a compatible kernel using Buildroot or Linux Kernel sources
  • A week for troubleshooting!

The following diagram outlines the overall process of IoT device emulation.

Adding a new device

Steps involved:

  1. Copy the template directory to make a new device configuration.
  2. Compile a matching kernel from source, and place it in the kernel/ directory.
  3. Copy the extracted rootfs from the device's firmware into the rootfs/ directory. Typically these would be SquashFS or CramFS filesystems, uncompressed using binwalk or unsquashfs or cramfsck.
  4. Place the contents of extracted nvram in nvram.ini
  5. Edit the config file with the newly populated device firmware contents.
  6. Create a new device record in the devices file. Pay close attention to QEMU command line options.

The following sample kernels are provided with the template.

  • zImage-2.6.39.4-vexpress ARMv7 CPU on a vexpress-a9 board.
  • zImage-2.6.31.14-realview-rv130-nothumb ARMv6 CPU on a realview-eb board.
  • zImage-2.6.31-versatile-nothumb ARMv5 CPU on a versatilepb board.

However, it is encouraged to build a compatible kernel from source.

ARM-X Presentations

Presentation at Countermeasure 2019 on 7 November 2019.

<iframe src="//www.slideshare.net/slideshow/embed_code/key/6P5quK19YMwYQ5" width="595" height="485" frameborder="0" marginwidth="0" marginheight="0" scrolling="no" style="border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;" allowfullscreen> </iframe>

Release presentation at HITB+Cyberweek on 16 October 2019.

<iframe src="https://www.slideshare.net/slideshow/embed_code/key/9FqUwLVZaoLaxO" width="595" height="485" frameborder="0" marginwidth="0" marginheight="0" scrolling="no" style="border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;" allowfullscreen> </iframe>

The ARM IoT Firmware Laboratory - NEW TRAINING

An all new class where the ARM IoT EXPLOIT LABORATORY leaves off. The ARM IoT Firmware Laboratory dives into analysis, extraction and emulation of IoT device firmware, using a variety of techniques. Students shall be given ample hands on practice in emulating a variety of IoT devices. Lab exercises feature firmware extraction directly from the hardware, building a custom kernel and buildroot environment, extracting contents of nvram and emulating the device under ARM-X. The class also goes on to fuzzing and exploit development exercises for the emulated devices.

Upcoming classes:

  1. HITB2020AMS, Amsterdam: (3 day class) https://conference.hitb.org/hitbsecconf2020ams/sessions/3-day-training-1-the-arm-iot-laboratory/

  2. Ringzer0 2020, Las Vegas: (4 day class) https://ringzer0.training/arm-iot-firmwarelab.html

Downloads

Pre-built VM with ARM-X installed

VMware VM: https://app.box.com/s/3iyi5f6vpakngh8ti3ir2zzukgdu0j2q

The ARM-X VM is compressed using 7-Zip. The archive is split into multiple files of 200MB each, because several cloud hosting providers impose a maximum limit. To extract the VM, use the 7z command line utility:

7z e armx-march2020.7z.001

SHA 256 Checksums:

286fbfcda02f5161d4fc983584873ccd5462c315a8c9783232cbef717c5c054e  armx-march2020.7z.001
f513145ed451bc3e07014824deb9ffa5c66b0ebad97ba24aed500abcdea6f7f6  armx-march2020.7z.002

1a41f4a028db80787adbeac03bfb20794994f3658e67764c200d885498521d2f  armx-mar2020.vmx
eba62de7c6235859e2150503a29f4bf84a85dd9dba5c5ba46e44940de6a03387  armx-s001.vmdk
5feeb87965d0e04dff31bbc6c61eb676e1a8c72c11076ba2167d960195391dd1  armx.vmdk

VirtualBox VM: (coming soon, but don't hold your breath)

ARM-X Code

Github: https://github.com/therealsaumil/armx/

ARM-X Documentation

Tutorial: Debugging With ARM-X

END

ARM-X is licensed under the Mozilla Public License v2.0 (MPLv2).

  • v0.9.0 22-October-2019, Preview Release
  • v0.9.1 19-November-2019
  • v0.9.2 12-March-2020

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.