首页 | 本学科首页   官方微博 | 高级检索  
     

一种6读2写多端口寄存器堆的全定制实现
引用本文:张轩,李兆麟.一种6读2写多端口寄存器堆的全定制实现[J].计算机工程,2007,33(20):248-250.
作者姓名:张轩  李兆麟
作者单位:清华大学信息科学技术学院,CPU & SoC实验室,北京,100084
摘    要:采用全定制设计方法实现了一种6读2写的3232位的多端口寄存器堆,包括结构设计、电路设计、版图设计、仿真验证以及建模建库。该多端口寄存器堆的读写端口互相独立,在一个时钟周期内,能够同时读出6个32位数据,并写入2个32位数据。在电路实现上,采用高速SCL结构的地址译码和分组字线的方法来减少读写延迟。采用了0.18µm 6层金属P阱CMOS工艺来实现版图设计,通过了版图验证和后端仿真。

关 键 词:多端口寄存器堆  高速SCL结构  全定制设计  自顶向下的设计流程
文章编号:1000-3428(2007)20-0248-03
修稿时间:2006年11月19

Full-custom Implementation of 6-read 2-write Multi-port Register File
ZHANG Xuan,LI Zhao-lin.Full-custom Implementation of 6-read 2-write Multi-port Register File[J].Computer Engineering,2007,33(20):248-250.
Authors:ZHANG Xuan  LI Zhao-lin
Affiliation:(CPU & SoC Lab, Information Science and Technology College, Tsinghua University, Beijing 100084)
Abstract:This paper presents a full-custom design of a multi-port 32x32 bits register file with 6 read ports and 2 write ports, including architecture design, circuit design, layout design, simulation and verification, modeling and library compiling. As all read/write ports are independent, it can read 6 32-bit data and write 2 32-bit data simultaneously in one clock cycle. The design employs high-speed SCL address-decoder and word-line grouping method to reduce access delay. Its layout is realized in 0.18μm 6-layer-metal P-trap CMOS technology and passes final verification and post simulation.
Keywords:multi-port register file  high-speed SCL structure  full-custom design  top-down design flow
本文献已被 万方数据 等数据库收录!
点击此处可从《计算机工程》浏览原始摘要信息
点击此处可从《计算机工程》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号