Provided are a floating-point unit, a system, and method for fused multiply-add logic to process input operands including floating-point values and integer values. A first input operand comprising an integer value and second and third input operands comprising floating-point values are received. The first, second, and third input operands are processed to produce a floating-point result.