$ go version
go version go1.14.1 linux/amd64

The documentation for regexp package says that "All characters are UTF-8-encoded code points". However, it doesn't say anything about what happens if the input is not valid UTF-8.

The current behavior on invalid UTF-8 is not consistent. For example, when matching against the string "\xff", the pattern a doesn't match, the pattern \x{fffd} doesn't match either, but the pattern a|\x{fffd} surprisingly does match.

The documentation should be updated to specify the behavior of regexp package on invalid UTF-8 input. Also, the behavior should be made more consistent (for example, always convert every undecodeable byte into \ufffd).