“Cube attack” less effective against Trivium than we thought?

It looks like there are errors in the tables at the back of the “cube attack” paper which show how to apply the attack to Trivium: some of the entries don’t work. This could mean simply that there are typos in the table, or it could mean that the attack is somewhat less effective against…

Asynchronous libraries performance

Recently I found some pretty libevent benchmarks. For me they show terrifying results. The blood freezing fact is that the more connections you have, the bigger is the cost of adding new connections to asynchronous loop. It means that if you have 1 connection registered to asynchronous loop, the cost of registering callback would be…

Trivium, SSE2, CorePy, and the “cube attack”

I present a new implementation of the stream cipher Trivium designed for cryptanalysts, in particular those interested in applying the "cube attack" to Trivium. It generates 128 simultaneous output streams using SSE2 intrinsics, and achieves under 1 cycle/byte, over four times faster than standard implementations. The entire program is in Python; SSE2 machine instructions are generated and called using the tool CorePy, an approach I am happy to recommend to others with similar needs. The code is under the MIT licence and may be found in this Mercurial repository.