Carry Save Array Multiplier for Dual Arithmetic Operations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional processors require dedicated circuitry for arithmetic operations, leading to increased cost and complexity, especially for multiply and accumulate (MAC) instructions, which are not optimized for performance and cost balance.

Innovation Solution

A processor design incorporating a carry save array multiplier with cascaded partial product generators and sequencing logic that can perform both multiplication and addition, eliminating the need for a dedicated adder for double precision results, thereby reducing circuit area and cost while maintaining performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If dedicated circuitry is provided for arithmetic operations including separate adder and multiplier, then arithmetic functionality and reliability are improved, but device complexity and manufacturing cost increase

Engineering Contradiction:
Improvearithmetic functionalityVSAvoidcircuit complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The carry save array multiplier is designed to perform both multiplication and addition operations using the same hardware circuitry. The multiplier can be configured to execute multiplication when control signals indicate multiply operations, and addition when control signals indicate add operations, eliminating the need for separate dedicated adder and multiplier circuits while maintaining full arithmetic functionality

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent combines the functions of separate adder and multiplier circuits into a single carry save array multiplier unit. By merging these previously distinct arithmetic operations into one unified hardware structure, the overall device complexity and component count are reduced while preserving all necessary arithmetic capabilities

Inventive Principle:
Principle #5Merging (Combining)

2Manufacturing precision

If separate adder circuitry is provided for double precision results, then manufacturing precision is improved, but device complexity and area increase

Engineering Contradiction:
Improvedouble precision arithmeticVSAvoidcircuit area
Core Design Contradiction:
Manufacturing precisionVSArea of stationary object

Solution Approach 1:

The carry save array multiplier serves dual purposes by handling both single-precision multiplication and double-precision addition operations. When double precision results are required, the same multiplier circuitry is utilized to perform the necessary addition operations, eliminating the need for separate dedicated adder circuitry and thereby reducing the overall circuit area while maintaining double precision accuracy

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If conventional hardware multiplier design is used, then arithmetic performance is improved, but device complexity and cost increase

Engineering Contradiction:
Improvearithmetic performanceVSAvoidcircuit complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The multiplier circuit incorporates dynamic control mechanisms that allow it to adapt its operation mode based on the required arithmetic function. Control signals dynamically configure the carry save array to perform either multiplication or addition, enabling the circuit to maintain high arithmetic performance for different operations while using a single reconfigurable structure rather than multiple static dedicated circuits

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS10929101B2Processor with efficient arithmetic units
Publication Date: 2021.02.23 TEXAS INSTRUMENTS INC
  • US10929101B2 patent drawing
  • US10929101B2 patent drawing
  • US10929101B2 patent drawing

AI summary

A processor includes a carry save array multiplier. The carry save array multiplier includes an array of cascaded partial product generators. The array of cascaded partial product generators is configured to generate an output value as a product of two operands presented at inputs of the multiplier. The array of cascaded partial product generators is also configured to generate an output value as a sum of two operands presented at inputs of the multiplier.