Using a more efficient string matching algorithm

It does, it's just slow enough to emulate that (given how heavily it's used in memchr) that it's an immediate red flag to me.

Popping up a level though, I don't think this is a blocker? We could just have the acceleration for x86.

Well, it's quite unfortunate, but yeah, it's not a blocker.

Yeah happy to have an issue on memchr about this. This issue might be a good place actually?

Thanks, I'll try to bring it up over the weekend.