Hi,
> It seems JIT is 16 times!! slower than interpreter for such simple pattern.
I did some improvements on the SSE2 accelerated search and /(?s).*/ search. You can try them now. However I have never seen such big differences in my measurements. The memchr can use AVX2, so it is still faster, but it is not clear how you can detect AVX2 at runtime. At least this fails on Ubuntu:
