https://en.wikipedia.org/wiki/Endianness#Floating_point

However, on modern standard computers (i.e., implementing IEEE 754), one may in practice safely assume that the endianness is the same for floating-point numbers as for integers, making the conversion straightforward regardless of data type

igbinary_unserialize_double/igbinary_serialize_double

Filing this to document it, though no bugs were reported

/** Serializes double. */
inline static int igbinary_serialize_double(struct igbinary_serialize_data *igsd, double d) {
	union {
		double d;
		uint64_t u;
	} u;

	u.d = d;

	return igbinary_serialize8_and_64(igsd, igbinary_type_double, u.u);
}