Faster version of reference decoder

[hoehrmann]

Hi, I read your http://nullprogram.com/blog/2017/10/06/ - interesting work. For what it's worth, note that at the end of http://bjoern.hoehrmann.de/utf-8/decoder/dfa/#variations there is an improved version of the decoder that saves a shift for every byte, compared to the original version at the beginning of the article. Might be interesting to use the improved version as reference instead.

[skeeto]

You're right, the improved version is much faster! In the same setup (with the same benchmark that clearly favors the branchless version) it's 10% slower with GCC than the branchless version (improved from 20% slower). However, with Clang it's 20% _faster_ than the branchless version. I'll update my article to mention this. Thanks for getting in touch!