US 5,689,721 AGranted

Detecting overflow conditions for negative quotients in nonrestoring two's complement division

AssigneeCyrix Corporation

InventorRobert D. Maher, III

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method of detecting anomalous overflow conditions is used, in an exemplary embodiment, in implementing in a 486-type microprocessor, nonrestoring two's complement division for negative quotients using 2n bit dividends and n bit divisors. Each interative division step, an adder/subtractor is used to add/subtract the properly aligned divisor to/from the left shifted dividend, to produce a partial remainder and a carry out bit Cout. Complement Cout is assumed to be the same as the most significant bit of the partial remainder PR(MSB), such that PR(MSB) is used as the sign bit in further computations, with complement Cout being used to control quotient generation according to DVRS XOR Cout. The anomalous overflow test signals overflow when complement Cout is the different than the most significant bit of the first partial remainder PR1(MSB), such that the anomalous overflow test is implemented according to the logic equation: Cout XNOR PR1(MSB).

Claims8

Dependent claims

In a system with iterative divide circuitry for performing nonrestoring division using two's complement 2n bit signed dividends and two's complement n bit signed divisors, overflow detection circuitry for detecting overflow conditions specifically in cases that yield two's complement negative quotients, comprising:

iterative divide circuitry that receives a signed dividend and a signed divisor of opposite signs, when a resulting quotient is negative;
and
first and second overflow detection circuitry coupled to the iterative divide circuitry;
in a first iterative step, the iterative divide circuitry shifts out a sign bit of the signed dividend, and then adds the signed divisor, properly aligned, to obtain a first partial remainder, and a carry out bit which is designated Cout;
the first overflow detection circuitry providing in response to the first partial remainder and Cout a first overflow signal if the magnitude of the first partial remainder causes an overflow condition to occur for any n-bit divisor;
if the first overflow detection circuitry does not provide the first overflow signal, the iterative divide circuitry is operable (i) to perform a sequence of iterative divide steps using successive partial remainders and the signed divisor to obtain a sequence of quotient digits and successive partial remainders, and then (ii) to provide an adjusted signed quotient;
the second overflow detection circuitry being responsive to the adjusted signed quotient to provide a second overflow signal if the adjusted signed quotient cannot be stored in two's complement format in an n bit register;
whereby the first and second overflow signals identify all cases of overflow.

5 more elements

iterative divide means for receiving a signed dividend and a signed divisor of opposite signs, when the quotient is negative;
in a first iterative step, the iterative divide means shifting out a sign bit of the dividend, and then adding the divisor, properly aligned, to obtain a first partial remainder and a carry out bit designated Cout;
first overflow detection means coupled to the iterative divide means for providing in response to the first partial remainder and Cout a first overflow signal if the magnitude of the first partial remainder causes overflow to occur for any n-bit divisor;
if the first overflow detection means does not provide the first overflow signal, the iterative divide means is operable (i) to perform a sequence of iterative divide steps using successive partial remainders and the signed divisor to obtain a sequence of quotient digits and successive partial remainders, and then (ii) to provide an adjusted signed quotient;
second overflow detection means for providing, responsive to the adjusted signed quotient, a second overflow signal if the adjusted signed quotient cannot be stored in twos complement format in an n bit register;
whereby the first and second overflow signals identify all cases of overflow.

3 more elements

Dependencies are inferred from the claim text.