Multiply add functional unit capable of executing scale, round, getexp, round, getmant, reduce, range and class instructions
A method is described that involves executing a first instruction with a functional unit. The first instruction is a multiply-add instruction. The method further includes executing a second instruction with the functional unit. The second instruction is a round instruction.
1. A processor comprising:
a decode unit to decode a first instruction, the first instruction to indicate a first floating point term X and a second term Y; and
an execution unit to execute the first instruction, the execution unit including:
at least an exponent calculation logic and a mantissa calculation logic, to scale the first floating point term X by an integer associated with the second term Y to generate a floating point result R, wherein the floating point result R is equal to the first floating point term X multiplied by two raised to a power of the integer associated with the second term Y; and
write logic circuitry to store the floating point result R.
2. The processor of claim 1 , wherein a mantissa of the floating point result R is equal to a mantissa of the first floating point term X, and an exponent of the floating point result R is equal to an exponent of the first floating point term X plus the integer associated with the second term Y.
3. The processor of claim 1 , wherein the floating point result R has a binary point of the first floating point term X moved by the integer associated with the second term Y.
4. The processor of claim 1 , wherein the execution unit comprises a masking layer.
5. The processor of claim 2 , wherein the floating point result R is part of a packed operand, and comprising a masking layer to present arithmetic exceptions that occur for valid floating point results of the packed operand and is to suppress presenting arithmetic exceptions that occur for floating point results of the packed operand that are not valid.
6. The processor of claim 1 , wherein the first floating point term X is a scalar single precision floating point term.
7. The processor of claim 1 , wherein the first floating point term X is a scalar double precision floating point term.
8. The processor of claim 1 , wherein the first floating point term X is one of a plurality of single precision floating point terms in a packed operand.
9. The processor of claim 1 , wherein the first floating point term X is one of a plurality of double precision floating point terms in a packed operand.
10. The processor of claim 1 , wherein the decode unit and the execution unit are disposed on a semiconductor chip.
11. The processor of claim 1 , wherein the processor is a Complex Instruction Set (CISC) architecture processor.