cemu

cemu-linux

Description

Writing assembly is fun. Assembly is the lowest language (humanly understandable) available to communicate with computers, and is crucial to understand the internal mechanisms of any machine. Unfortunately, setting up an environment to write, compile and run assembly for various architectures (x86, ARM, MIPS, SPARC) has always been painful. CEmu is an attempt to fix this by providing a bundled GUI application that empowers users to write assembly and test it by compiling it to bytecode and executing it in an QEMU-based emulator.

Cheap EMUlator combines all the advantages of a basic assembly IDE, compilation and execution environment, by relying on the great libraries Keystone, Unicorn and Capstone engines in a Qt powered GUI.

It allows to test binary samples, check your shellcodes or even simply learn how to write assembly code, all of this for the following architectures:

x86-32 / x86-64
Arm / AArch64
MIPS / MIPS64
SPARC / SPARC64
(more to come)

CEmu was tested and works on Linux, Windows and MacOSX.

Installation

Quick install with PIP

Notes:

if you are using Kali Linux, there is a known problem with the installation of the keystone-engine package using PIP. A quick'n dirty fix for that would be (as root):

$ sudo updatedb
$ sudo locate libkeystone.so
$ sudo ln -sf /path/to/libkeystone.so/found/above /usr/local/lib/libkeystone.so

if you are using OSX, there is also a known issue when installing capstone-engine from PIP, resulting in the .dylib not being deployed at the right location. A quick fix for it is

# locate the shared lib
$ find ~  -type f -name libcapstone.dylib
# link it in a valid correct library path
$ ln -sf /path/to/libcapstone.dylib/found/above /usr/local/Cellar/python3/3.6.2/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/capstone/libcapstone.dylib

From PyPI

This is the recommended way to install cemu as it will work out of the box. You can install cemu on your system or using virtualenv or pipenv, by running:

pip3 install --user --upgrade cemu

From GitHub

git clone https://github.com/hugsy/cemu && cd cemu
pip3 install --user --upgrade .

Manual installation

If for some reason the installation via PIP fails, you can always run cemu by installing manually the following dependencies:

unicorn and its Python bindings, as the emulation engine
keystone and its Python bindings, as the assembly engine
capstone and its Python bindings, as the disassembly engine
PyQt5 for the GUI
pygments for the text colorization

Linux / OSX

Use the script requirements.sh to install all the dependencies required for cemu.

Since some packages can be installed via your package manager, the script may ask for your root password if required.

$ ./requirements.sh

By default, the script will install the dependencies to have cemu running with Python3. If you prefer to use Python2, simply add --python2 to the command line, like this:

$ ./requirements.sh --python2

Windows

The fastest way for Windows is to install the packaged binaries for: * Keystone (http://www.keystone-engine.org/download/) * Capstone (http://www.capstone-engine.org/download) * Unicorn (http://www.unicorn-engine.org/download/)

Then spawn cmd.exe and install the missing Python packages: python-qt5, pygments).

C:>pip.exe install python-qt5 pygments

If you are running Python2, you will also need to install the package enum34.

OpenReil integration

`OpenREIL <https://github.com/Cr4sh/openreil>`__ is an Open Source library created by [@Cr4sh](https://twitter.com/@d_olex) that implements a translator and tools for generating Intermediate Language level code (REIL). OpenREIL library can be used optionally with cemu. The Current version of OpenREIL only provides support for x86 (Intel) architecture.

If you use cemu with Python 2.7, you can also use `OpenReil <https://github.com/Cr4sh/openreil>`__ to generate IR code based on the content of the Code panel.

cemu-openreil

To do so, follow the installation procedure for OpenReil by following the steps here.

Contribution

cemu was created and maintained by myself, `@_hugsy_ <https://twitter.com/_hugsy_>`__, but kept fresh thanks to all the contributors.

abstractmachines/cemu