logo资料库

Introduction to VLSI Systems.pdf

第1页 / 共423页
第2页 / 共423页
第3页 / 共423页
第4页 / 共423页
第5页 / 共423页
第6页 / 共423页
第7页 / 共423页
第8页 / 共423页
资料共423页,剩余部分请下载后查看
Introduction to VLSI Systems
Preface
Background
Contents
1. MOS Devices and Circuits
2. Integrated System Fabrication
3. Data and Control Flow in Systematic Structures
4.Implementing Integrated System Designs
5.Overview of an LSI Computer System; Design of the OM2 Data Path Chip
6.Architecture and Design of System Controllers
7. System Timing
8.Highly Concurrent Systems
9.Physics of Computational Systems
、. -.. V*'. » • » •• • > 二 -■ • I nRJ • v .. >、•:;^u运 m>、 ,f v»»«;•_ - - ... nT juui i K JnHti ^. ^vrrrrvl o^il ■声----- O-K > r^npl.hl - - - - V_— M__r fll ypy i r. r i> 声,,尸尸尸产 — ^ 卢In t 友 被~£ ;i ^ r 轉^' :- 1*1®圖ilL [ 厂: II : i i s r n k v ^ H K i ^ S J P — 發 ^5a 、、_T ^l^X ;/■ '.lllLyi.idt ■^rEBL 一 「f f If jall - liii lBI si M ; .£*■ : . .. 1 Ififi .^iy iMILJ 11B.. iSiil4 SB !ly .. t i1- - - 1 i l ! ■ l • i l l i l ^ L W j y r p “ 尹: 蛛 兵 口 Z 二一 *• 1 - 1 ; JLT^ST^S5v々 IV,立 • ^,4C 4^. *A rrsr^4^-i r^r^lLr、為:^A ^ p ^ r z ^ 一 .«1 1 二> 二 ‘i 》 . , (Hut r. -ajfj^ t ^ I m .m _ i ’ j g ' l a i [ , _ W $ S ~ ^ - . f e ■ 碎 链 , - 德 ^ - t e s : • ' - 鐵 戾 1 祐 綠 约 i f e f e u v ! ::- , ' ^ 3 MhHRH ^Nhu^ *w ?r U j!r馨 ppjiJ4“嘴* 1
A硪MsooS2A1 lnloJtv< Jse!J}l SJyuslsoais.^UTAca CMDs/a^ puv -Mspl OOJ J-iov 妁v tfC & io^< .llslvjodjou AN
◎1. i L . _ _ _ _ l £al>e4o a« l a定 s souJo s . C CUU IVOIJOJBJulmas J.yf >UJ ~^U‘LSZ 's.?cs p^advpls.. SBOJ一ao 益 ololilpyl fosmiisul.
This book is in the Ac3di^on-V/os!ey Series in Computer Science Consulting Editor Michael A- Harrison Library of Congress Cataloging in Publication Data Mead, Carver A Introduction to V LS I systems. 1- Integrated circuits— Large scale integration. 2. Microcomputers. 4. Computer architecture. II. Title, joint amiior, TK7874.M37 _ ISBN 0-701-04358-0 3. Digital electronics. I* Conway,Lynn A ., 621.381^535 73-7468S Second printing, October 1980 O Copyright ◎ 1980 by Addison-Wesley Publishing Company„ Inc. Philippines copynght i9f>0 by Addison-Wcsley Publishing Company, [ac. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmiltcd, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior wnuen permission of the publisher. Printed in the United States of America. Published simtiltaneoutsly in Canada, Library of Congress Catalog Card No. 78-74688. ISBN 0-?01-0435S-0 FGHIJKLMN-KA-8站765432
TO W. R_ SUTHEHLAND
A W ..:..'...:.-. PREFACE What we have seen so far is only the beginning. Achievable circuit density now doubles with each passing year or two. Physical principles indicate that transistors can be scaled down to less than 1/100th of their present area and still function as the sort of switching elements with which we can build digital systems. By the late 1980s it will be possible to fabricate chip's containing millions of transistors. The devices and interconnections in such very large scale integrated (VLSI) systems will have linear dimensions smaller than the wavelength of visible light. New high-resolution lithographic techniques have already been demon­ strated that will enable fabrication of such circuitry. VLSI electronics presents a challenge, not only to those involved in the development of fabrication technology, but also to computer scientists and com­ puter architects. The ways in which digital systems are structured, the procedures used to design them, the trade-offs between hardware and software, and the design of computational algorithms will all be greatly affected by the coming changes in integrated electronics. We believe this will be a major area of activity in computer science on through the 1980s. As a result of improvements in fabrication technology, Large Scale fntegrated (LSI) eieclronic circuitry has become so dense that a single silicon LSI chip may contain tens of thousands of transistors. Many LSI chips, such as microproces­ sors, now consist of multiple complex subsystems, and thus are really integrated systems rather than integrated circuits. Until recently the design of integrated circuitry has been the province of circuit and logic designers working within semiconductor firms. Computer ar- diitects. have traditiorrally composed systems from standard integrated circuits. designed and manufactured by these firms but have seldom participated in the specification and design of these circuits. Electrical Engineering and Computer Science (EE/CS) curricula reflect this tradition, with courses in device physics and integrated circuit design aimed at a different group of students than those interested in digital system architecture and computer science. , - ........... r r f j J s •'-;. w ‘• ^ ■ ^
-------------- ----------1 ---------- ' ......................................................... ..............................................- ' ' ” - - -------------- -------- 一 . . - k d vj Preface This text is written to fill a current gap in the literature and to introduce all EE/CS students to integrated system architecture and design. Combined with individual study in related research areas and participation in large system design projects, ihis text provides the basis for a graduate course-sequence in integrated systems. However, it is primarily intended for use in intensive undergraduate courses on the subject. The material can also be used to augment courses on computer architecture. We assume the reader's background contains the equiva­ lent of introductory courses in computer science, electronic circuits, and digital design. There have been major obstacles in the way of those seeking an overall understanding of integrated systems. Integrated electronics has developed in a heatedly competitive and often secretive business environment. There lias been a proliferation of different device technologies, circuit design families, logic design techniques, maskmaking and wafer fabrication techniques, etc. Many of these technologies have sprung up from the grass roots of “ Silicon Valley,,in the San Francisco Bay Area of California, and thus many of the “ experts” are located in that one region. Most workers in the industry have concentrated on narrow specialties. Separate integrated electronics cultures have independently evolved within many companies, and thus the terminology and practices of the specialties vary from company to company. As a result of this background, texts on integrated electronics have tended to give detailed accounts of some very narrow horizontal segment of the overall subject, such as device physics or circuit design, and are often tied in subtle ways to some specific context, thus limiting their general applicability. We have chosen instead to provide just enough essential information about devices,circuits, fabrication technology, logic design techniques, and system architecture to enable the reader to fully span the entire range of abstractions from the underlying physics to complete VLSI digital computer systems. A rather small set of key concepts is sufficient. Only by learning the essence of each topic, and by carrying along the least amount of mental baggage at each step, w ill the student emerge with a good overall understanding of the subject. This understanding can then be mapped into the reader's own space of application, technology, and technical culture. The high rate of change oHntegrated electronics presents another obstacle: information often becomcs obsolete very rapidly. The major force for obsoles- cence is the ongoing improvement in fabrication technology, leading to smaller and smaller devices as time passes and thus to a constant change in device characteristics. We attack this obstacle by stressing the effects of the scaling- down of device dimensions. Many of the coming changes in system architectural parameters are thus anticipated. The reader will learn what is common to systems composed of 6 弘m,2 jam, and 0.5 devices, and what is not. While the material in this text is presented in a particular order, it need not be read in that order. Each chapter presents material from a distinct level in the ,% 、'. :泛 • 夂 、 , 0 , r:. ,'v •' .I一P p . . I f . 響 : .I : w p l p 7, p v*-. f . . ^ m v ' - w w^.'7- . I'-..I看1 r 'v , , f7 ' ?§ ! ' I ^ 。 ... T 1 . . . - r? ( - —
Preface hierarchy of disciplines involved in integrated systems. The material falls into four major groupings: Chapters 2 and 2 provide the basics of devices, circuits, and fabrication; Chapters 3 and 4 give the basics of system design and implementation; Chapters 5 and 6 present an example of LSI system design; topics of current research interest are discussed in Chapters 7., 8,and 9. We recommend that readers start in the chapter where they are most knowledgeable, and read until information is required from an adjacent area described in some other chapter. By using this algorithm and consulting the suggested references where necessary, readers can gradually work through the primary material of all chapters. Although much of the materia! in this text is previously unpublished, it nevertheless con­ tains only basic concepts. However, these concepts cover quite a wide range of disciplines and are easily visualized only after the overall context becomes clear. In any given technology, form follows function in a particular way. The most efficient first step towards understanding the architectural possibilities of a technology is the study o f carefully selected existing designs. However, system architecture and design,like any art, can only be learned by doing. Carrying a small design from conception through to successful completion provides the confidence necessary to undertake larger designs. The space of possibilities un­ folds only as the medium is worked. This book provides a set of selected design examples and also describes procedures for implementing one's own designs. Because of the density, speed, and topological properties of «MOS, and the easy access to /jMOS wafer fabrication, that technology is used for our examples. The architectural skill of mapping function into form, when once acquired, can then be extended to other technologies. The geneml availability o f courses in VLSI system design at major universi­ ties marks the beginning o f a new era in electronics. The rate of systems innova­ tion using this re markable technology need no longer be limited by the perceptions of a handful of semiconductor companies .and large computer manufacturers. New metaphors for computation, new design methodologies, and an abundance of new application areas are already arising within the universities, within many system firms, and within a multitude o f new small enterprises. There may never have been a greater opportunity fo r free enterprise than that presented by these circum­ stances. An atmosphere of excitement and anticipation pervades this field. A growing community of workers from many backgrounds, computer scientists, clectrical engineers, mathematicians, and physicists are collaborating on a common prob­ lem area that has not yet become classical. The territory is vast, and largely unexplored. The rewards are great for those who simply press forward. Pasadena, California Belmont, Californio July 1979 C.M. L.C. " r ' ' . r-- * i r rri r Ir ” 'i- -;_7V^. ^ wr; ^!;^ ^ t '-vr. ^ -A‘<> ‘; i> . Kt■■ 《_ . . • ; L. V
分享到:
收藏