System and method of floating point multiply operation processing
AssigneeQUALCOMM Incorporated
InventorsAlbert Danysh, Erich Plondke, Eric Mahurin
Life of this patent
Estimating the term…
Abstract
A processor includes an integer multiplier configured to execute an integer multiply instruction to multiply significand bits of at least one floating point operand of a floating point multiply operation. The processor also includes a floating point multiplier configured to execute a special purpose floating point multiply accumulate instruction with respect to an intermediate result of the floating point multiply operation and the at least one floating point operand to generate a final floating point multiplication result.
Claims20
A processor comprising:
- an integer multiplier configured to execute an integer multiply instruction to multiply significand bits of at least one floating point operand of a floating point multiply operation;
- and
- a floating point multiplier configured to execute a special purpose floating point multiply accumulate instruction with respect to an intermediate result of the floating point multiply operation and the at least one floating point operand to generate a final floating point multiplication result.
A method of performing a floating point multiply operation at a processor, the method comprising:
- executing a plurality of integer multiply instructions to generate an intermediate result;
- and
- executing a special purpose floating point multiply accumulate instruction with respect to the intermediate result and at least one floating point operand of the floating point multiply operation to generate a floating point multiplication result.
An apparatus comprising:
- means for executing a plurality of integer multiply instructions to generate an intermediate result;
- and
- means for executing a special purpose floating point multiply accumulate instruction with respect to the intermediate result and at least one floating point operand to generate a final floating point multiplication result.
A method comprising:
- identifying a first operand of a multi-operand multiply operation as denormal;
- and
- in response to determining that an exponent of a second operand of the multi-operand operation is larger than a threshold, modifying a value of at least one of the first operand or the second operand.
Dependencies are inferred from the claim text.