Investigate ternary removal in UTF-8 decoder
Andy Wingo mentioned today about some work in Chromium which removed the ternary from the UTF-8 decoder (which is based on the same one that VTE uses).
https://chromium-review.googlesource.com/c/v8/v8/+/805357
We might be able to do the same and pick up a bit more performance in PTY throughput.
Also mentioned:
might want to see if utf8-decoder.h in the dir from the OP gives you any benefit. note the caveat on the commit log: "This places a requirement on the caller to 0 out the code point between successful decodings"