US 5,710,730 AGranted

Divide to integer

AssigneeInternational Business Machines Corporation

InventorRonald Morton Smith, Sr.

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system and method for providing an interruptible remainder instruction that can produce a quotient as well as a remainder. Remainders are computed through an iterative procedure. This procedure is carried out in a computer system's hardware by following a series of steps, the series being interruptible at any point. Each step reduces the magnitude of the dividend until the final remainder can be obtained. In the intermediate steps, the sign of the new (smaller in magnitude) dividend is kept the same as the sign of the original dividend, and the value Ni (which can be considered part of the quotient) is rounded toward zero. Only in the last step must the sign of the operands be considered and directed rounding be performed. Throughout the remainder operation, the partial quotients can be saved so that upon completion, not only has the remainder been computed, but so has the quotient.

Claims11

Dependent claims

In a computer system having a floating point processor, a method for computing an integer quotient and a remainder resulting from the division of a floating point dividend by a divisor, said integer quotient representing the integer value nearest an infinitely exact quotient of said dividend divided by said divisor, said dividend having a number of original dividend digits, comprising the steps of:

a) using said floating point processor to compute a first value equal to the quantity of individual digits contained in a quotient represented by said dividend divided by said divisor;
b) using said first value and said floating point processor to compute a second value equal to the quantity of individual digits in a partial quotient, said second value having a maximum value equal to a predetermined partial quotient fraction number of digits;
c) using said floating point processor to divide the dividend by the divisor to provide said partial quotient having a quantity of digits equal to said second value, and to provide a partial remainder;
d) storing said partial quotient in a memory;
e) assigning a new value to the dividend, said new value being equivalent to said partial remainder;
and
f) repeating steps a through e until all of said original dividend digits have been used to form partial quotients, thereby providing the floating point quotient and the floating point remainder.

4 more elements

In a computer system having a floating point processor, a method for computing an integer quotient and a remainder resulting from the division of a floating point dividend by a floating point divisor, said integer quotient representing the integer value nearest an infinitely exact quotient of said dividend divided by said divisor, said remainder representing the difference between said dividend and a product formed between said integer quotient and said divisor, said dividend and said divisor each represented in a floating point format having a significand of a predetermined number of bits, said integer quotient represented by one or more partial fractions each having a maximum number of bits equal to a predetermined partial-quotient fraction number having a value not greater than said predetermined number of bits of said significand, said method comprising the steps of:

a) generating a signal representing a total count of how many digits are in a quotient that represents the integer value of an infinitely exact quotient of said dividend divided by said divisor;
b) generating a signal representing a first count of individual digits for a first partial quotient, said first count equal to a quantity represented by said total count of digits in said integer quotient portion modulo said partial-quotient fraction number when said quantity is non-zero, said first count equal to said partial-quotient fraction number when said quantity is zero;
c) generating said partial quotient by dividing said dividend by said divisor to provide a number of digits equal to said first count;
d) generating a partial remainder signal equal to a difference between said dividend and a product formed between said divisor and said partial quotient, each digit of said partial quotient weighted by its respective significance;
e) repeating steps a through d, with said partial remainder substituted for said dividend, until all digits of said integer quotient have been generated, corresponding to repeating steps a through d until said total count is not greater than said partial-quotient fraction number, thereby providing said integer quotient and said remainder represented by said partial quotients and said partial remainder, respectively.

2 more elements

Dependencies are inferred from the claim text.