// Float32bits returns the IEEE 754 binary representation of f, // with the sign bit of f and the result in the same bit position. // Float32bits(Float32frombits(x)) == x. func Float32bits(f float32) uint32 { … } // Float32frombits returns the floating-point number corresponding // to the IEEE 754 binary representation b, with the sign bit of b // and the result in the same bit position. // Float32frombits(Float32bits(x)) == x. func Float32frombits(b uint32) float32 { … } // Float64bits returns the IEEE 754 binary representation of f, // with the sign bit of f and the result in the same bit position, // and Float64bits(Float64frombits(x)) == x. func Float64bits(f float64) uint64 { … } // Float64frombits returns the floating-point number corresponding // to the IEEE 754 binary representation b, with the sign bit of b // and the result in the same bit position. // Float64frombits(Float64bits(x)) == x. func Float64frombits(b uint64) float64 { … }