Pipelined packet-forwarding floating point. I. Foundations and a rounder

作者: D.C. Matula , A.M. Nielsen

DOI: 10.1109/ARITH.1997.614889

关键词:

摘要: The paper presents the foundations for a packet forwarding floating point format and design of rounder ensuring compatibility between standard binary IEEE 754 format. related addition multiplication algorithms described in this series propose new ALU pipeline paradigm handling data hazards pipelined operations. execution phases adder multiplier pipelines are illustrated by proposed implementation having four stages. latter two stages each employ herein. phase intended to map logic designs, with only some fifteen levels per stage allowing be mapped reasonably short cycles. provides input output cycle effective latency cooperating pipelines. designs we cut half reduce stall cycles factor three compared conventional processing dependent speedup is realized, preservation compatibility.

参考文章(6)
Amos R. Omondi, Computer arithmetic systems: algorithms, architecture and implementation Prentice Hall International (UK) Ltd.. ,(1994)
Naofumi Takagi, Michael J. Flynn, Nhon Quach, On fast IEEE rounding Stanford University. ,(1991)
M.R. Santoro, G. Bewick, M.A. Horowitz, Rounding algorithms for IEEE multipliers symposium on computer arithmetic. pp. 176- 183 ,(1989) , 10.1109/ARITH.1989.72824
R.K. Yu, G.B. Zyner, 167 MHz radix-4 floating point multiplier Proceedings of the 12th Symposium on Computer Arithmetic. pp. 149- 154 ,(1995) , 10.1109/ARITH.1995.465364
A.M. Nielsen, D.W. Matula, C.N. Lyu, G. Even, Pipelined packet-forwarding floating point. II. An adder symposium on computer arithmetic. pp. 148- 155 ,(1997) , 10.1109/ARITH.1997.614890
S.F. Oberman, M.J. Flynn, Design issues in division and other floating-point operations IEEE Transactions on Computers. ,vol. 46, pp. 154- 161 ,(1997) , 10.1109/12.565590