HELP in MIPS

[hawaii86]

Hi, every one
MIPS assembly language program to do single-precision floating-point multiplication in software rather than in hardware. The procedure float_multiply should receive its input parameters in $a0 and $a1 (as single-precision floating-point numbers) and produce its result in $v0 (as single-precision float). You cannot use the floating-point multiply instruction mul.s to do the multiplication. Only integer instructions are allowed. Write additional procedures if needed to extract and compute the result exponent and the result significand.
You should also make sure to handle special cases:
Multiplication by zero, infinity, and NaN
Overflow and underflow
Denormalized numbers