Questions about math::cstdfloat/boost::float128_t

Hello,
please see https://godbolt.org/z/3qMfehf6z

It seems that the intel compiler supports several functions:
- complex: __cXq
- strtoflt128/quadmath_snprintf

With Linux it works. My questions:
1) Are the functions also available on Windows (intel compiler) ?
2) Are the workarounds for cygwin still needed ?
3) Are there any other restrictions, e.g. regarding compiler versions (boost::math needs C++14 and therefore up-to-date compilers) ?

I can't check this myself, because I have neither Windows nor intel compiler.

thx
Gero