How to cast a 32bit float to 64bit integer without relying on static_cast or 64-bit floating point operations?

Question

I need to convert an IEEE 754 32bit float into a 64bit signed int (rounding to nearest even) without going through a static_cast or triggering 64bit fl