Generally, I wanted to find out more about the Linux Kernel and this was a nice side project.
Speed. Speed. Speed.
Seriously, coming as close as possible to the performance of fizzbuzz. This is a highly optimized assembler version of the fizzbuzz interview question. It uses VLIWs and specifically the L2 cache buffer size.
Checkout the commit history to see the progress.
I followed this excellent blog post.
I put the fizzbuzz assembler file in this repo, but I do not claim ownership nor license it.