Design issues and implementations for floating-point divide-add fused

Authors:
Alexandru Amaricai;Mircea Vladutiu;Oana Boncalo
Affiliations:
Advanced Computing Systems and Architectures Research Group, University Politehnica of Timisoara, Timisoara, Romania;Advanced Computing Systems and Architectures Research Group, University Politehnica of Timisoara, Timisoara, Romania;Advanced Computing Systems and Architectures Research Group, University Politehnica of Timisoara, Timisoara, Romania
Venue:
IEEE Transactions on Circuits and Systems II: Express Briefs
Year:
2010

Citing 16
Cited 0

Interval analysis for computer graphics

SIGGRAPH '92 Proceedings of the 19th annual conference on Computer graphics and interactive techniques
Area and performance tradeoffs in floating-point divide and square-root implementations

ACM Computing Surveys (CSUR)
Design Issues in Division and Other Floating-Point Operations

IEEE Transactions on Computers
On-the-Fly Rounding (Computing Arithmetic)

IEEE Transactions on Computers
167 MHz Radix-8 Divide and Square Root Using Overlapped Radix-2 Stages

ARITH '95 Proceedings of the 12th Symposium on Computer Arithmetic
Delay-Optimized Implementation of IEEE Floating-Point Addition

IEEE Transactions on Computers
Digit Selection for SRT Division and Square Root

IEEE Transactions on Computers
Digit-Recurrence Dividers with Reduced Logical Depth

IEEE Transactions on Computers
Floating-Point Fused Multiply-Add: Reduced Latency for Floating-Point Addition

ARITH '05 Proceedings of the 17th IEEE Symposium on Computer Arithmetic
Computing machine-efficient polynomial approximations

ACM Transactions on Mathematical Software (TOMS)
Towards more efficient interval analysis: corner forms and a remainder interval newton method

Towards more efficient interval analysis: corner forms and a remainder interval newton method
Design of A Double-Precision Floating- Point Multiply-Add-Fused Unit with Consideration of Data Dependence

ISVLSI '07 Proceedings of the IEEE Computer Society Annual Symposium on VLSI
Design of the ARM VFP11 Divide and Square Root Synthesisable Macrocell

ARITH '07 Proceedings of the 18th IEEE Symposium on Computer Arithmetic
PCU: the programmable culling unit

ACM SIGGRAPH 2007 papers
Verified Real Number Calculations: A Library for Interval Arithmetic

IEEE Transactions on Computers
Design of Low-Cost High-Performance Floating-Point Fused Multiply-Add with Reduced Power

VLSID '10 Proceedings of the 2010 23rd International Conference on VLSI Design

Quantified Score

Hi-index	0.00

Visualization

Abstract

This brief presents a dedicated unit for the combined operation of floating-point (FP) division followed by addition/subtraction--the divide-add fused (DAF). The goal of this unit is to increase the performance and the accuracy of applications where this combined operation is frequent, such as the interval Newton's method or the polynomial approximation. The proposed DAF unit presents a similar architecture to the FP multiply-accumulate units. The main difference is represented by the divider, which is implemented using digit-recurrence algorithms. An important design tradeoff regarding DAF is represented by the number of required quotient bits. We present the impact of the adopted number of quotient bits on accuracy, cost, and performance. Consequently, two implementations are proposed: one pro-accuracy and one pro-performance. We show that the proposed implementations have better accuracy with respect to the solution based on two distinct units: an FP divider and an FP adder. The implementation suitable for lower latency presents the best cost-performance tradeoff.