26.1.13 Floating-point arithmetic and relational instructions

The following instructions assume that both operands have the same size in bits.