[eresi-dev] A new cycle of ERESI (important message to read)

[jv at ens dot fr]

Hello ERESI coders,

Its been a little while we havent been in contact all together
for various reasons (around 1 month now). I think everyone
finished his exams and everything as we are September 16.

I have myself some time now to conclude what has been done
for the last 6 months. I hope everyone can participate in the
upcoming month to stabilize the project for the release !

Many of you have been involved in ERESI. Some from the
beginning, but most of you for 2 years, 1 year, and even
for some of you, only more recently. It is time for everyone
of you to choose if you continue on ERESI for a new cycle
of features or if you decide to stop your involvement now.

I believe ERESI is a project that will grow in the future. The
package has just started to exist, after 6 years of elfsh-centric
development. It has been time for us to get more open and
become the initiators of this new community project.

ERESI also already started to have more academical recognition,
by our desire to involve it in our university cursus. For example:

* Julio Auto has done a succesful Bachelor project in the Federal
University of Pernambuco in Brazil  by implementing in ERESI
language the backend for the 8086 instruction set to the ERESI
low-level intermediate form (ELIR).

* I have been developping the ERESI language as being
a research intern in the University of Cambridge
Computer Laboratory in the UK as a master thesis of
research in computer science of the University of Paris.

Our work is based on the ERESI language, which permits
program transformation and data-flow analysis of binary programs.
The Evarista analyzer is almost entirely written in the ERESI
language. The work of Julio Auto and I is integrated in Evarista,
as such is a part of the ERESI framework.

Evarista is still in development and we need manpower to port it
to other architectures (for exemple: MIPS). But we also have integrated
succesfully a new project in the ERESI framework : kernsh. 

Kernsh starts a new cycle of kernel-level features within the ERESI 
framework.
It has been developed by Anthony Desnos on top of ERESI after a first 
standalone
version by Samuet Dralet and Nicolas Brito. For now, kernsh allows to 
script the
kernel memory inside the ERESI language for tasks such as code 
injection, memory
allocation, inspection, modification and more). Kernsh comes to complete 
elfsh and
e2dbg by bringing static and runtime kernel facilities. Kernsh is 
currently available
only for the Linux kernel.

More ERESI activity has been realized or planed:

* Anthony Desnos and Julio Auto exhibed a voluntee to continue
on ERESI and make it a master project. I think its a great
idea. I am also thinking about ERESI for my Phd topic about
program analysis. In other words, ERESI is a project that has a
great academic potential in the next years and I encourage everyone
to allocate scholar time for thinking and developping ERESI, for
instance as a final year project in your degree. The ERESI team
brings great support to new developer and this is an opportunity
for you to start being serious with program analysis and reverse
engineering, an increasingly popular topic in research and the
industry.

* Latest Phrack #64 featured an article about the vision leading
to developping the Evarista analyzer. "Automated vulnerability
auditing in machine code" explains how Chevarista, the ancestor
of Evarista, written as an IDA plugin, has been showing great
potential and how we now start to implement such analyzer in
our own ERESI framework.

For all these reasons, it is important now to know who else
desires to continue or to start with ERESI. If you are one of
these persons, answer to this email directly on the list.

- Julien Vanegue (for the ERESI team)