For a moment now, I've been wanting to try the intel AVX instructions. So I decided to write an SHA256 function in pure x86-64 assembly. That algorithm might not be the best thing to parallelize though but it still was fun to do.
Update dec 16 2015: I have modified my code to use AVX2 instructions. I recently bought a Intel i5-6400 which supports a lot of new instructions I didn't have before. So I modified the algorithm to use bleeding-edge instructions.
The use of AVX instructions in this algorithm might not give better performances. The only reason I did this was because I wanted to play with AVX. Using AVX2 probably helps a lot more since a lot of the AVX instructions are eliminated.
In fact, I'm not entirely convinced that using AVX will benefit a lot. One thing to consider here, is that for small hashes it could be slower. The thing is that when using AVX, you are using the xmm/ymm registers. During a context switch, the OS does not automatically save/restore the state of the AVX registers. Those are lazy-saved. Meaning that, without going too much into the details, the CPU will only save/restore the AVX registers if it detects that the current thread is using them (using a dirty flag and an exception). Such a save/restore is crazy expensive. So introducing the usage of AVX registers in a thread will cost a lot for the context switch, yielding less processing time for the thread. So the thing to consider is: will the thread use the AVX instructions enough to overcome the cost of the context switch?