FLOATING-POINT ERROR DETECTION AND CORRECTION

摘要

An embodiment includes a method for detecting a potential floating-point error in an addition or a subtraction instruction included in an operation. The method may include identifying a first operand and a second operand. The first operand and the second operand may be configured to be manipulated during execution of the instruction. The method may include copying a first exponent of the first operand to a first comparison register. The method may also include copying a second exponent of the second operand to a second comparison register. The method may further include comparing the first exponent in the first comparison register to the second exponent in the second comparison register. Based on the comparison, a determination may be made whether the instruction includes a potential floating-point error when executing the instruction using the first operand and the second operand formatted according to a first precision.

主权项

1. A computer readable media including computer executable instructions configured to cause a processing device to perform operations for correcting potential floating-point errors in an addition or a subtraction calculation, the operations comprising:
receiving a first subset of a plurality of instructions within an original program, the first subset including a first instruction; for the first instruction, copying a first exponent of a first operand into a first register and copying a second exponent of a second operand into a second register; comparing the first exponent to the second exponent to generate an exponent comparison; and when the exponent comparison is above a first preselected threshold or below a second preselected threshold, calculating a result of the first instruction using the first operand formatted according to an increased precision and the second operand formatted according to the increased precision.