ebook img

Microprocessor architecture : from simple pipelines to chip multiprocessors PDF

384 Pages·2010·13.072 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Microprocessor architecture : from simple pipelines to chip multiprocessors

This page intentionally left blank MICROPROCESSOR ARCHITECTURE This book gives a comprehensive description of the architecture of microprocessorsfromsimplein-ordershortpipelinedesignstoout-of- ordersuperscalars.Itdiscussestopicssuchas (cid:1) Thepoliciesandmechanismsneededforout-of-orderprocessing, suchasregisterrenaming,reservationstations,andreorderbuffers (cid:1) Optimizations for high performance, such as branch predictors, instructionscheduling,andload–storespeculations (cid:1) Designchoicesandenhancementstotoleratelatencyinthecache hierarchyofsingleandmultipleprocessors (cid:1) State-of-the-artmultithreadingandmultiprocessing,emphasizing single-chipimplementations Topics are presented as conceptual ideas, with metrics to assess the effects on performance, if appropriate, and examples of realization. The emphasis is on how things work at a black box and algorithmic level. The author also provides sufficient detail at the register trans- fer level so that readers can appreciate how design features enhance performanceaswellascomplexity. Jean-LoupBaerisProfessorEmeritusofComputerScienceandEngi- neeringattheUniversityofWashington,wherehehasbeensince1969. ProfessorBaeristheauthorofComputerSystemsArchitectureandof morethan100refereedpapers.HeisaGuggenheimFellow,anACM Fellow,andanIEEEFellow.Baerhasheldseveraleditorialpositions, includingeditor-in-chiefoftheJournalofVLSIandComputerSystems andeditoroftheIEEETransactionsonComputers,theIEEETrans- actionsonParallelandDistributedSystems,andtheJournalofParallel andDistributedComputing.HehasservedasGeneralChairandPro- gramChairofseveralconferences,includingISCAandHPCA. Microprocessor Architecture FROM SIMPLE PIPELINES TO CHIP MULTIPROCESSORS Jean-Loup Baer UniversityofWashington,Seattle CAMBRIDGEUNIVERSITYPRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Dubai, Tokyo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521769921 © Jean-Loup Baer 2010 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2010 ISBN-13 978-0-511-67546-1 eBook (NetLibrary) ISBN-13 978-0-521-76992-1 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. ToDiane,Marc,Shawn,andDanielle Contents Preface pagexi 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 AQuickViewofTechnologicalAdvances 2 1.2 PerformanceMetrics 6 1.3 PerformanceEvaluation 12 1.4 Summary 22 1.5 FurtherReadingandBibliographicalNotes 23 exercises 24 references 28 2 TheBasics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29 2.1 Pipelining 29 2.2 Caches 46 2.3 VirtualMemoryandPaging 59 2.4 Summary 68 2.5 FurtherReadingandBibliographicalNotes 68 exercises 69 references 73 3 SuperscalarProcessors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75 3.1 FromScalartoSuperscalarProcessors 75 3.2 OverviewoftheInstructionPipelineoftheDECAlpha21164 78 3.3 IntroducingRegisterRenaming,ReorderBuffer,and ReservationStations 89 3.4 OverviewofthePentiumP6Microarchitecture 102 3.5 VLIW/EPICProcessors 111 3.6 Summary 121 3.7 FurtherReadingandBibliographicalNotes 122 exercises 124 references 126 vii viii Contents 4 Front-End:BranchPrediction,InstructionFetching,andRegister Renaming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 4.1 BranchPrediction 130 Sidebar:TheDECAlpha21264BranchPredictor 157 4.2 InstructionFetching 158 4.3 Decoding 164 4.4 RegisterRenaming(aSecondLook) 165 4.5 Summary 170 4.6 FurtherReadingandBibliographicalNotes 170 exercises 171 ProgrammingProjects 174 references 174 5 Back-End:InstructionScheduling,MemoryAccessInstructions, andClusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177 5.1 InstructionIssueandScheduling(WakeupandSelect) 178 5.2 Memory-AccessingInstructions 184 5.3 Back-EndOptimizations 195 5.4 Summary 203 5.5 FurtherReadingandBibliographicalNotes 204 exercises 205 ProgrammingProject 206 references 206 6 TheCacheHierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 6.1 ImprovingAccesstoL1Caches 209 6.2 HidingMemoryLatencies 218 6.3 DesignIssuesforLargeHigher-LevelCaches 232 6.4 MainMemory 245 6.5 Summary 253 6.6 FurtherReadingandBibliographicalNotes 254 exercises 255 ProgrammingProjects 257 references 258 7 Multiprocessors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 7.1 MultiprocessorOrganization 261 7.2 CacheCoherence 269 7.3 Synchronization 281 7.4 RelaxedMemoryModels 290 7.5 MultimediaInstructionSetExtensions 294 7.6 Summary 296 7.7 FurtherReadingandBibliographicalNotes 297 exercises 298 references 300

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.