Extended precision integer divide algorithm

摘要

A method for an extended precision integer divide algorithm. The method of one embodiment comprises separating a first L bits wide integer dividend into two equal width portions, wherein a first integer format portion comprises lower M bits of the first integer dividend and a second integer format portion comprises upper M bits of the first integer dividend, wherein M is equal to ½ L. The first integer format portion is converted into a first floating point format portion. An N bits wide integer divisor is converted from an integer format into a floating point format divisor. The first floating point format portion is divided by the floating point format divisor to obtain a first floating point format quotient. The first floating point format quotient is converted into a first integer format quotient. The second integer format portion is converted into a second floating point format portion. The second floating point format portion is divided by the floating point format divisor to obtain a second floating point format quotient. The second floating point format quotient is converted to a second integer format quotient. The first and second integer format quotients are summed together to generate a third integer format quotient.