The problem with that is that Unicode string processing is highly data dependent. You may not get good speedups because most performance improvement from puttin

The problem with that is that Unicode string processing is highly data dependent. You may not get good speedups because most performance improvement from putting things in hardware comes from parallelism in the data path. A good approach to getting speedups is probably by assuming ascii, vectorizing the hell out of that, and falling back to multibyte processing where that fails. Checking for multibyte characters comes down to checking if the high bit of any byte is set, which should be fast and easy for the branch predictor to deal with.