Data processor a method and apparatus for performing postnormalization in a floating-point execution unit
AssigneeMotorola, Inc.
InventorsDavid R. Bearden, Raymond L. Vargas
Life of this patent
Estimating the term…
Loading patent…
AssigneeMotorola, Inc.
InventorsDavid R. Bearden, Raymond L. Vargas
Estimating the term…
A method and apparatus for performing prenormalization during execution by an execution unit (100) of a floating-point add/subtract operation using two data operands. The execution unit (100) adds a mantissa portion of a first and a second floating-point data operand to generate a prenormalized mantissa sum. The execution unit (100) minimizes critical path delays to allow high-performance floating-point calculations while simultaneously reducing logic. Instead of treating the prenormalized mantissa sum as a 64-bit value with special treatment in case of a carry out due to overflow, the floating-point adder 100 treats the prenormalized mantissa sum as a 65-bit value, with the most significant bit being a carry output. Instead of conditionally incrementing an initial exponent value, the initial exponent value is always incremented. Thus, allowing the floating-point adder unit 100 to perform the exponent adjustments for normalization and for rounding faster.
In a data processor having a register file for storing a plurality of floating-point data operands, each of the floating-point data operands having a mantissa portion, an exponent portion and a sign bit, and an execution unit for executing a plurality of floating-point operations using a selected number of the plurality of floating-point data operands, a method for performing a floating-point addition/subtraction operation to determine a final floating-point result, the method comprising the steps of:
In a data processor having a register file for storing a plurality of floating-point data operands, each of said floating-point data operands having a mantissa portion, an exponent portion and a sign bit, and an execution unit for executing a plurality of floating-point operations using a selected number of said plurality of floating-point data operands, a method for performing postnormalization during a floating-point addition/subtraction operation to determine a final floating-point result, said method comprising the steps of:
A data processor having a register file, for storing a plurality of floating-point data operands, where each of said floating-point data operand has a mantissa portion, an exponent portion, and a sign bit, and a floating-point execution unit for performing a floating-point addition/subtraction operation to determine a floating-point result, said execution unit comprising:
Dependencies are inferred from the claim text.