The AU-A7540 Floating Point Core provides IEEE754-1985 compliant single and double
precision floating point. It implements floating point addition, subtraction, multiplication,
division, conversions, comparisons, negation, and absolute value in hardware. IEEE754-1985
compliant results are completely computed by hardware. The Floating Point Core is available as
a synthesizable Verilog model from Aurora VLSI, Inc.

• IEEE754-1985 floating point
• Fully compliant results for all floating point operations are computed in hardware
• Single and double precision
• Floating point operations include:
- addition- single and double precision
- subtraction- single and double precision
- multiplication- single and double precision
- division- single and double precision
- float to float conversions- single to double, double to single
- float to integer conversions- single/double to 32/64 bit integer
- integer to float conversions- 32/64 bit integer to single/double
- comparision- single and double precision
- negation- single and double precision
- absolute value- single and double precision
• Rounding modes
- round to nearest
- round to zero
- round to positive infinity
- round to negative infinity
• Exceptions detected- inexact, overflow, underflow, invalid, and divide by zero