US 5,218,564 AGranted

Layout efficient 32-bit shifter/register with 16-bit interface

AssigneeNational Semiconductor Corporation

InventorsRalph W. Haines, Gary D. Phillips, D. Kevin Covey, Thomas W. S. Thomson

Life of this patent

Estimating the term…

Loading patent…

Abstract

An integrated circuit processor architecture that implements digital signal processing (DSP) functions with less hardware, improved speed and a more efficient layout. The central processing unit (CPU) resources are used in conjunction with an integrated multiply/accumulate unit to perform DSP operations. Use of the CPU's internal register for the circular buffer of the DSP multiply/accumulate function allows a minimum amount of lower speed hardware to be used for the multiply/accumulate unit and permits DSP operations to be performed in parallel. The multiply/accumulate unit takes advantage of the inherent accumulating properties of conventional multiplier designs to perform multiplication of two signed binary numbers using the modified Booth's algorithm but in both reduced cycle time and hardware requirements. This is accomplished by using the adder within the multiplier to sum the product terms. Instead of clearing the adder of the result of one multiplication before beginning another, the result is maintained and all subsequent partial products are added to it to generate a final output. The placement of the 32-bits of the multiply/accumulate unit's multiplicand register is arranged in two rows of 8 even bits and two rows of 8 odd bits to allow left shift by two with a single loop around and to provide direct interface with the 16-bit data latch register on its input and the rest of the 32-bit arithmetic logic unit (ALU) on its output.

Claims1

Dependent claims

A multiply/accumulate unit for multiplying and accumulating binary data, the multiply/accumulate unit being physically formed as part of a monolithic integrated circuit die, the multiply/accumulate unit comprising:

(a) a multiplier register for storing a 16-bit binary multiplier value;
(b) a multiplicand input register for storing 16 bits of a 32-bit binary multiplicand value;
(c) a multiplicand register connected to the multiplicand input register for storing a 32-bit binary multiplicand value received by the multiplicand register via the multiplicand input register;
(d) an arithmetic unit connected to the multiplier register and to the multiplicand register for multiplying the binary multiplier value received from the multiplier register and the binary multiplicand value received from the multiplicand register to provide a 32-bit product;
and
(e) a 32-bit accumulator register connected to the arithmetic unit for storing the product received from the arithmetic unit
wherein the multiplicand register is physically arranged in four rows of 8 bits each, the top row containing the even bits for the low word, the second row containing the even bits for the high word, the third row containing the odd bits for the low word, and the fourth row containing the odd bits for the high word,
whereby the arrangement of the multiplicand register allows left shift by two directly across with a single loop around between bits 14-16 and 15-17 and
whereby the 16-bit input register fits exactly across the top of a multiplicand stack.

6 more elements

Dependencies are inferred from the claim text.