Release of my code contribution to engine-cuda

[Johannes Gilger]

Hey Paolo (and everyone still reading this),

it has finally happened, my paper (spinoff from my Diploma Thesis) was
accepted at a conference in September of this year:
http://web.sec.uni-passau.de/isc2012/ (Sept 19-21, Passau, Germany).

This also means that I can finally release my source code, which I did
about five minutes ago: https://github.com/heipei/engine-cuda

I've tried to document everything as good as possible, but there are
bound to be some problems here and there, especially when mixing
different GCC/CUDA-Versions and GPUs. I don't really have the time to
address these issues at the moment, so your mails to this list
complaining about something might fall on deaf ears for now.

The repository includes the complete history with a few patches by Paolo
in between, so you can easily see what I did compared to the original
code base. As a short summary, here is what is different:

* CUDA support for DES, Blowfish, Camellia-128, CAST-5, IDEA
All algorithms supported in ECB mode, CBC decryption

* Key-scheduling done on the CPU

* Improved (coarse-grained) version of AES

* engine-opencl with AES, DES, BF, CMLL, CAST5, IDEA

* Test-scripts, autoconf adjustments
Preprocessor macros to time the kernel execution time

* Lots of restructuring, heavy macro use

Have fun with it!

Regards,
Jojo

--
Johannes Gilger <hei...@hackvalue.de>
http://heipei.net
GPG-Key: 0xD47A7FFC
GPG-Fingerprint: 5441 D425 6D4A BD33 B580 618C 3CDC C4D0 D47A 7FFC

[Paolo Margara]

Hi Johannes,
I'm happy for you that the article taken from your thesis has been
accepted and you've decided, finally, to publish the code that you have
produced during it's writing.
I would have preferred that your code was published directly into the
project's repository, not into another one.
Depending by my time availability I will proceed to merge the new
repository within the pre-existing one, that is the official one of the
project, for now I'll put a link on the home of the project at the
repository you have given to me.
I hope this does not mean the beginning of a fork of the project,
because this would not be a good thing for a software that has been
essentially developed by only two developers who have released code
publicly in the last two years.

Regards,
Paolo Margara

[Johannes Gilger]

On 18/07/12 16:10, Paolo Margara wrote:
> I would have preferred that your code was published directly into the
> project's repository, not into another one.

Hi Paolo,

as I've been developing using git I've decided to simply upload my
git-repo to github, since it was much easier for me. Also I think that
it will attract more contributors than on Google Code, which is kind of
read-only through the use of SVN.

My intention was never to fork engine-cuda, merely to extend it. What we
could do is that you also put engine-cuda on your GitHub-account and
that I mark my own repo as a fork of your repo (so that people will
recognize you as "upstream").

I probably wont have much time to continue working on engine-cuda, so
having you as the upstream maintainer is probably the best idea.

[Johannes Gilger]

A quick follow-up, for those of you that are interested: My paper on my
engine-cuda improvements is available online.

Springer: http://www.springerlink.com/content/y4uxq2734xx52416/
Draft-Version: http://itsec.rwth-aachen.de/publications/engine-cuda.pdf

Maybe I'll also put up my diploma thesis soon. Too bad all the interest
in engine-cuda seems to have died down :(