DOCSLIB.ORG

Procesory Ve Směrovačích Firmy Cisco Motorola MPC857DSL

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Procesory Ve Směrovačích Firmy Cisco Motorola MPC857DSL Procesory ve směrovačích firmy Cisco Pokročilé architektury počítačů Marek Malysz, mal341 Obsah Motorola MPC857DSL.............................................................................................................................1 Motorola 68360.........................................................................................................................................2 Motorola 68030.........................................................................................................................................3 Motorola MPC860 PowerQUICC............................................................................................................3 PMC-Sierra RM7061A.............................................................................................................................4 Broadcom BCM1250................................................................................................................................5 R4600.........................................................................................................................................................5 R5000.........................................................................................................................................................6 R7000.........................................................................................................................................................6 QuantumFlow Processor...........................................................................................................................6 Závěr..........................................................................................................................................................7 Zdroje.........................................................................................................................................................8 Úvod Firma Cisco Systems patří mezi přední výrobce síťových zařízení. V jejich portfoliu nalezneme širokou škálu síťových řešení - hardwarových i softwarových. Jejich zařízení najdou uplatnění v intrastruktůře menších podníků, ale taky v nejnáročnější podmínkách poskytovatelů internetového připojení nebo datových center. Je tedy nedílnou součástí používat procesory optimalizované pro síťové komunikace a zajistit co nejmenší zdržení při odbavování paketů. Tato práce si klade za cíl vytvořit přehled procesorů, které používá firma Cisco ve svých směrovačích. Jsou zde zahrnuty i již nepodporované typy směrovačů. Seřazeny jsou podle řady, do kterých přináleží. Motorola MPC857DSL Jedná se o všestranný mikroprocesor s podporou vnějších rozhraní, který najdeme nejen v řídících systémech, ale zejména v zařízeních na straně zákazníka nebo i DSL modemech. Obsahuje 32- bitové MPC8xx jádro s Power architekturou. Pracuje na frekvencích 50-66MHz a obsahuje sdílenou cache paměť o velikosti 8KB. Je tvořen dvěmí částmi: MPC8xx jádrem a pomocným procesorem pro komunikaci (Communication Processor Module - CPM). CPM podporuje čtyři seriove řadiče (serial communications controllers - SCCs) na zařízení. Tato dvou-procesorová architektura umožnila snížit spotřebu el. energie díky odlehčení samotného jádra o periferní zátěže na úkor CPM. Koprocesor pro řízení komunikace podporuje operace na FastEthernetových i sériových portech. Ilustrace 1: Blokový diagram procesoru MPC857DSL Procesor byl použit v směrovačích Cisco 83x series. 1 Motorola 68360 Jedná se o nástupce procesoru MC68302, který obsahuje 4 sériové řadiče. Primární procesor je 32- bitový CISC pracující na frekvenci 25-33 MHz. Ilustrace 2: Blokový diagram procesoru MC68360 1 http://www.cisco.com/en/US/docs/ios/12_2/12_2y/release/notes/rn800yn.html Procesor byl použit v směrovačích Cisco 1600 series. 2 Ilustrace 3: Blokový diagram směrovače 1600 Motorola 68030 První verze tohoto procesoru vznikla v roce 1998 jako další z řady 680x0. Jedná se v podstatě o jádro z 68020 s přídanou datovou pamětí a technologii umožňující vměstnat více hardwaru do jednoho čipu. Mikroprocesor je 32 bitový s architekturou CISC. Obsahuje dvě oddělené L1 cache paměti, datovou a instrukční, každá má 256 bytů. Dále obsahuje vlastní jednotku správy paměti s podporou stránkování (PMMU), je schopný adresovat až 4GB paměti. Pro její správu byla instrukční sada rozšířena o další 4 instrukce. Navíc obsahuje patici pro dodatečné připojení jednotky pro operace v plavoucí řadové čárce (FPU). Šířka datové sběrnice je dynamická 8, 16 a 32bitová. Procesory byly vyráběny s frekvencemi od 15 do 50 MHz. Tyto procesory jsou využivány v směrovačích Cisco 2500 series. Taktovány na frekvencích 20-25 MHz.3 Ilustrace 4: Blokové schéma směrovače řady 2500 Motorola MPC860 PowerQUICC Jedná se o všestranný mikroprocesor s podporou vnějších rozhraní, který najdeme nejen v řídících systémech ale zejména v komunikačních a síťových zařízeních. Jedná se o další generaci procesoru MC68360 QUICC. Jeho název vychází z ¹quadª (4 podporovaní seriová rozhraní) Integrated Communication Controller. Stejně jako jeho předchůdce je tvořen dvěmí částmi: MPC8xx jádrem a pomocným procesorem pro komunikaci (CPM). Tyto procesory obsahovaly 4-16Kb instrukční a 4-8Kb datové L1 cache paměti, každá s vlastní MMU. Sběrnice byla dynamická 8-32 bitů. Frekvence koprocesoru v byla v rozsahu 50-80 MHz. 2 http://www.cisco.com/en/US/products/hw/routers/ps214/prod_bulletin09186a0080091fec.html 3 https://supportwiki.cisco.com/ViewWiki/index.php/Cisco_2500_Series_Router_Architecture Ilustrace 5: Blokový diagram procesoru MPC860 Tyto procesory jsou využivány ve směrovačích Cisco 2600 series kromě verze 2691, která používala RM7061A.4 PMC-Sierra RM7061A Jedná se o symetrický superskalární mikroprocesor, který je schopen vydávat 2 instrukce v jednom taktu. Běží na frekvencích do 400MHz. Mikroprocesor je 64 bitový s architekturou MIPS. Do čipu jsou integrovány 16KB instrukční a 16KB datové L1 cache, a taky 256KB L2 cache. Násobení i sčítání čísel v plavoucí řadové čárce zvládne tento procesor spočítat v jediném taktu. Ilustrace 6: Blokový diagram procesoru RM7061A 4 https://supportwiki.cisco.com/ViewWiki/index.php/Cisco_2600_Series_Router_Architecture Tyto procesory jsou využivány ve směrovačích Cisco 2691, 3620 a 3640.5 6 Broadcom BCM1250 Jedná se o dvou-jádrovy 64-bitový MIPS procesor. Dva jádra Broadcom SB-1 běží každé na 600 - 1000 MHz a sdílejí společnou 512 KB L2 cache. Obsahují oddělěné paměti instrukční a datové, každá 32KB. Na chipu lze dále najít řadič paměti DDR i integrovaný I/O subsystém. Základní bloky jsou mezi procesory propojené sdílenou sběrnicí s rozdělenými transakcemi - Zbus. Do procesoru lze připojit 3 gigabitové konektory a 2 seriové porty. Ilustrace 7: Blokové schéma procesoru BCM1250 Tento procesor je využíván v směrovačích Cisco 7200 Series v routovacím modulu Network Processing Engine NPE-G1. Taktovány jsou na 700 MHz. Ilustrace 8: Blokové schéma směrovačů Cisco řady 7200 R4600 Procesor pocházející z firmy Quantum Effect Devices, Inc., kterou firma Cisco později koupila. 5 https://supportwiki.cisco.com/ViewWiki/index.php/Cisco_2600_Series_Router_Architecture 6 http://www.cisco.com/en/US/products/hw/routers/ps274/products_tech_note09186a00801e1155.shtml R4600 obsahuje instrukční sadu MIPS III s podporou pipeline s pěti stupni a 64 bitovou sběrnici. Obsahuje 16Kb instrukční and 16Kb datovou L1 cache. Byl vyráběn ve verzích od 100-135MHz s volitelnou 512kB L2 cache. Firma Cisco jej však na svých stránkách prezentuje s frekvencí 150 MHz. Tento procesor je využit ve směrovačích Cisco 7200 series v routovacích modulech NPE-100 a NPE-150.7 R5000 Tento superskalární procesor byl vyroben v roce 1996. Umožňuje vydávat 2 instrukce v jednom cyklu. Šířka sběrnice je 64bitů. Stejně jako R4600 podporuje pipeline s 5 stupni, navíc však obsahuje dvojnásobnou, tedy 32 kB instrukční a 32kB datovou L1 cache. Pracuje na frekvenci 262MHz. Tento procesor je využit ve směrovačích Cisco 7200 series v routovacím modulu NPE-200. R7000 Nástupce R5000. Obsahuje 2 jednotky FPU a 3 stupně cache paměti. Dvě L1 cache o velikosti 16KB a L2 cache 256KB jsou integrovány, třetí je volitelná. Tento procesor je využit ve směrovačích Cisco 7200 series v routovacích modulech NPE-300, NPE-400 a NSE-1. QuantumFlow Processor Jedná se o novinku z března roku 2008, kter firma uvedla na trh po 5-letém vývoji. Technologie na výrobu získala akvizicí 2 firem zajišťujících potřebné know-how. Procesor využívá architektury masivního multiprocessingu (MPP), obsahuje 40 32-bitových RISC jader pracujících na frekvenci 900-1200 MHz. Každé jádro obsahuje 4 vlákna mezi kterými sdílí 16KB L1 cache. L2 cache o velikosti 256KB je přístupná všem jádrům. Na výrobu první generace byla použitá technolgie 90 nm, v druhé generaci pak 65nm, čímž se podařilo vměstnat vše potřebné na jeden chip. Architektura se skládá z 2 hlavních částí: ● engine procesoru ± jež zajišťuje zpracování všech toků po dobu přeposílání paketu a ● traffic manageru ± zajišťuje řazení na frontách a plánování jak na systémových, tak na I/O rozhraních. 7 http://www.cisco.com/en/US/products/hw/routers/ps341/products_tech_note09186a0080094ea3.shtml Ilustrace 9: Blokové schéma procesoru QFC Procesor první generace je použit v ASR 1000, druhá generace pak v ASR 9000. Závěr Firma na svých oficiálních stránkách nezveřejňuje všechny používané procesory. Často jen informuje o výkonu použitého procesoru např 1.2 GHz PowerPC v případě ASR1000. Typ procesoru se přesto objevuje ve výpisu ¹show versionª na jednotlivých zařízeních. Viz příklad se zvýrazněným označením procesoru:
Load more

Recommended publications
  • 12 Questions To... Natami Team - Part 1
    12 questions to... Natami Team - part 1 Banter (c) Polski Portal Amigowy (www.ppa.pl) Today I will tell you about something unique. I will present you the latest 68k CPU! You may wonder... What actually is N68050? The N68050 (N050) is a new 68k family processor developed by the Natami Team. It runs inside the FPGA together with the Natami chipset. The 050 is the first revision of our 68k CPU architecture. Further on, we will extend the architecture to the N68050E and later the N68070 specification. The N68050 softcore will be the default CPU of the Natami system. But if the user wants to, he could also get a separate CPU card with a physical 68060 for his Natami system, connected to the mainboard over the SyncZorro bus. The development history of the N68050 core. One could say it started in 1998, when Gunnar tried to write his first 68k softcore. But FPGAs were unfortunately too small to accomplish this task back then. The actual development of the Natami softcore started two years ago, after we dismissed Coldfire as a possible CPU for the Natami. Coldfire was interesting, but our testing and estimates showed us that solving it with our own softcore is even better. The first "N68070" softcore design idea was similar to a crossbreed of 68000 and Coldfire. Our first concept was to use a pipeline similar to the Coldfire V3 pipeline. To improve performance we changed this and reworked the pipeline to be longer. Our goal for the softcore was to reach a higher clockrate and to be in all regards better than the original Motorola 68060 clocked at the same speed.
    [Show full text]
  • Lecture 1: Course Introduction G Course Organization G Historical Overview G Computer Organization G Why the MC68000? G Why Assembly Language?
    Lecture 1: Course introduction g Course organization g Historical overview g Computer organization g Why the MC68000? g Why assembly language? Microprocessor-based System Design 1 Ricardo Gutierrez-Osuna Wright State University Course organization g Grading Instructor n Exams Ricardo Gutierrez-Osuna g 1 midterm and 1 final Office: 401 Russ n Homework Tel:775-5120 g 4 problem sets (not graded) [email protected] n Quizzes http://www.cs.wright.edu/~rgutier g Biweekly Office hours: TBA n Laboratories g 5 Labs Teaching Assistant g Grading scheme Mohammed Tabrez Office: 339 Russ [email protected] Weight (%) Office hours: TBA Quizes 20 Laboratory 40 Midterm 20 Final Exam 20 Microprocessor-based System Design 2 Ricardo Gutierrez-Osuna Wright State University Course outline g Module I: Programming (8 lectures) g MC68000 architecture (2) g Assembly language (5) n Instruction and addressing modes (2) n Program control (1) n Subroutines (2) g C language (1) g Module II: Peripherals (9) g Exception processing (1) g Devices (6) n PI/T timer (2) n PI/T parallel port (2) n DUART serial port (1) g Memory and I/O interface (1) g Address decoding (2) Microprocessor-based System Design 3 Ricardo Gutierrez-Osuna Wright State University Brief history of computers GENERATION FEATURES MILESTONES YEAR NOTES Asia Minor, Abacus 3000BC Only replaced by paper and pencil Mech., Blaise Pascal, Pascaline 1642 Decimal addition (8 decimal figs) Early machines Electro- Charles Babbage Differential Engine 1823 Steam powered (3000BC-1945) mech. Herman Hollerith,
    [Show full text]
  • RTEMS CPU Supplement Documentation Release 4.11.3 ©Copyright 2016, RTEMS Project (Built 15Th February 2018)
    RTEMS CPU Supplement Documentation Release 4.11.3 ©Copyright 2016, RTEMS Project (built 15th February 2018) CONTENTS I RTEMS CPU Architecture Supplement1 1 Preface 5 2 Port Specific Information7 2.1 CPU Model Dependent Features...........................8 2.1.1 CPU Model Name...............................8 2.1.2 Floating Point Unit..............................8 2.2 Multilibs........................................9 2.3 Calling Conventions.................................. 10 2.3.1 Calling Mechanism.............................. 10 2.3.2 Register Usage................................. 10 2.3.3 Parameter Passing............................... 10 2.3.4 User-Provided Routines............................ 10 2.4 Memory Model..................................... 11 2.4.1 Flat Memory Model.............................. 11 2.5 Interrupt Processing.................................. 12 2.5.1 Vectoring of an Interrupt Handler...................... 12 2.5.2 Interrupt Levels................................ 12 2.5.3 Disabling of Interrupts by RTEMS...................... 12 2.6 Default Fatal Error Processing............................. 14 2.7 Symmetric Multiprocessing.............................. 15 2.8 Thread-Local Storage................................. 16 2.9 CPU counter...................................... 17 2.10 Interrupt Profiling................................... 18 2.11 Board Support Packages................................ 19 2.11.1 System Reset................................. 19 3 ARM Specific Information 21 3.1 CPU Model Dependent Features..........................
    [Show full text]
  • Miniaturization Technologies
    Chapter 1 Introduction and Summary “Small is Beautiful.” The truth of that state- technology is driven by a product or market dom- ment is debated in economic and sociological inated by another nation’s industry. circles, but when it comes to technology, there is no debate; small is beautiful because small is fast, The trends in silicon electronics miniaturiza- small is cheap, and small is profitable. The revo- tion show no signs of slowing in the near future. lution begun by electronics miniaturization dur- The current pace of miniaturization will pro- ing World War II is continuing to change the duce memory chips (dynamic random access world and has spawned a revolution in miniatur- memory, DRAM) with a billion transistors and ized sensors and micromechanical devices. the capacity to store 1 billion bits (1 gigabit) of in- Miniaturization plays a major role in the tech- formation around the year 2000.1 Transistors will nical and economic rivalry between the United continue to shrink until the smallest feature is States and its competitors. It translates to market around 0.1 micron (1 micrometer or one millionth share and competitive advantage for many com- of a meter). By comparison, today’s most ad- mercial and scientific products. Those compa- vanced mass-produced integrated circuits have nies and nations that can successfully develop features as small as 0.8 microns. A human hair is and capitalize on miniaturization developments 50 to 100 microns in width (see figure l-l). will reap handsome rewards. Personal comput- Achieving such tiny features will require a huge ers, portable radios, and camcorders are exam- engineering and research effort.
    [Show full text]
  • Modelagem Abstrata Para O Hardware De Mpsocs
    Pontifícia Universidade Católica do Rio Grande do Sul Faculdade de Informática Programa de Pós-Graduação em Ciência da Computação MODELAGEM ABSTRATA PARA O HARDWARE DE MPSOCS CARLOS ALBERTO PETRY Dissertação apresentada como requisito parcial à obtenção do grau de Mestre em Ciência da Computação na Pontifícia Universidade Católica do Rio Grande do Sul. Orientador Prof. Dr. Ney Laert Vilar Calazans Porto Alegre 2009 Dados Internacionais de Catalogação na Publicação (CIP) P498m Petry, Carlos Alberto Modelagem abstrata para o hardware de MPSoCS / Carlos Alberto Petry. – Porto Alegre, 2009. 113 p. Diss. (Mestrado) – Fac. de Informática, PUCRS Orientador: Prof. Dr. Ney Laert Vilar Calazans 1. Informática. 2. Multiprocessamento. 3. Modelagem de Sistemas. I. Calazans, Ney Laert Vilar. II. Título. CDD 004.35 Ficha Catalográfica elaborada pelo Setor de Tratamento da Informação da BC-PUCRS AGRADECIMENTOS Meu primeiro agradecimento é, com toda a certeza, ao divino Pai Eterno, Aquele que tudo provê e minha vida conduz. Em segundo lugar quero agradecer à minha família. À minha esposa pela paciência, pelo estímulo e pelas palavras carinhosas que sempre recebi. Aos meus filhos, por acreditarem e apoiarem meus sonhos, mesmo que isto tenha implicado, em muitas ocasiões, minha ausência em momentos de alegria, de tristeza e de necessidade. Agradeço também à minha família ascendente. À minha Mãe pela pessoa santa que é por dedicar toda a sua vida em prol de sua família, em especial a mim. Ao meu Pai, em memória, pelos exemplos de vida que me deixou. Aos meus irmãos José Alfredo, Ademar e Gelson, pelos grandes companheiros que tenho. E por fim à minha tia Lita por tantas orações que a mim dedicou.
    [Show full text]
  • Computer Architectures an Overview
    Computer Architectures An Overview PDF generated using the open source mwlib toolkit. See http://code.pediapress.com/ for more information. PDF generated at: Sat, 25 Feb 2012 22:35:32 UTC Contents Articles Microarchitecture 1 x86 7 PowerPC 23 IBM POWER 33 MIPS architecture 39 SPARC 57 ARM architecture 65 DEC Alpha 80 AlphaStation 92 AlphaServer 95 Very long instruction word 103 Instruction-level parallelism 107 Explicitly parallel instruction computing 108 References Article Sources and Contributors 111 Image Sources, Licenses and Contributors 113 Article Licenses License 114 Microarchitecture 1 Microarchitecture In computer engineering, microarchitecture (sometimes abbreviated to µarch or uarch), also called computer organization, is the way a given instruction set architecture (ISA) is implemented on a processor. A given ISA may be implemented with different microarchitectures.[1] Implementations might vary due to different goals of a given design or due to shifts in technology.[2] Computer architecture is the combination of microarchitecture and instruction set design. Relation to instruction set architecture The ISA is roughly the same as the programming model of a processor as seen by an assembly language programmer or compiler writer. The ISA includes the execution model, processor registers, address and data formats among other things. The Intel Core microarchitecture microarchitecture includes the constituent parts of the processor and how these interconnect and interoperate to implement the ISA. The microarchitecture of a machine is usually represented as (more or less detailed) diagrams that describe the interconnections of the various microarchitectural elements of the machine, which may be everything from single gates and registers, to complete arithmetic logic units (ALU)s and even larger elements.
    [Show full text]
  • Alocação De Tarefas E Comunicação Entre Tarefas Em Mps Ocs
    PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO GRANDE DO SUL FACULDADE DE INFORMÁTICA PROGRAMA DE PÓS-GRADUAÇÃO EM CIÊNCIA DA COMPUTAÇÃO ALOCAÇÃO DE TAREFAS E COMUNICAÇÃO ENTRE TAREFAS EM MPS OCS por CRISTIANE RAQUEL WOSZEZENKI Prof. Dr. Fernando Gehm Moraes Orientador Dissertação de mestrado submetida como requisito parcial à obtenção do grau de Mestre em Ciência da Computação. Porto Alegre, Março de 2007. Pontifícia Universidade Católica do Rio Grande do Sul Dados Internacionais de Catalogação na Publicação (CIP) W935a Woszezenki, Cristiane Raquel Alocação de tarefas e comunicação entre tarefas em MPSoCs / Cristiane Raquel Woszezenki. – Porto Alegre, 2006. 123 f. Diss. (Mestrado) – Fac. de Informática, PUCRS Orientador: Prof. Dr. Fernando Gehm Moraes 1. Multiprocessadores. 2. Alocação de Tarefas. 3. Arquitetura de Computadores. 4. Informática. I. Título. CDD 004.35 Ficha Catalográfica elaborada pelo Setor de Processamento Técnico da BC-PUCRS Campus Central Av. Ipiranga, 6681 – prédio 16 – CEP 90619-900 Porto Alegre – RS – Brasil Fone: +55 (51) 3320-3544 – Fax: +55 (51) 3320-3548 Email: [email protected] www.pucrs.br/biblioteca Resumo MPSoCs (do inglês, Multiprocessor System On Chip ) constituem uma tendência no projeto de sistemas embarcados, pois possibilitam o melhor atendimento dos requisitos da aplicação. Isso se deve ao fato de que a arquitetura desses sistemas é composta por vários processadores, módulos de hardware dedicados, memória e meio de interconexão, fornecendo um maior poder computacional quando comparados a sistemas monoprocessados equivalentes. No entanto, estratégias que possibilitem o aproveitamento da capacidade de processamento destas arquiteturas precisam ser mais bem entendidas e exploradas. Para isso, é necessário dispor de infra-estruturas de hardware e software que habilitem gerenciar a execução de tarefas no MPSoC.
    [Show full text]
  • Upgrade of Linac Control System with New Vme Controllers at Spring-8
    8th International Conference on Accelerator & Large Experimental Physics Control Systems, 2001, San Jose, California WEAT005 physics/0111131 UPGRADE OF LINAC CONTROL SYSTEM WITH NEW VME CONTROLLERS AT SPRING-8 T. Masuda, T. Ohata, T. Asaka, H. Dewa, T. Fukui, H. Hanaki, N. Hosoda, T. Kobayashi, M. Kodera, A. Mizuno, S. Nakajima, S. Suzuki, M. Takao, R. Tanaka, T. Taniuchi, Y. Taniuchi, H. Tomizawa, A. Yamashita, K. Yanagida SPring-8, Hyogo 679-5198, Japan Abstract In September 1999, we started to upgrade the present linac system to the new one by replacing a part of We integrated an injector linac control system to the hardware and introducing a standard software scheme as SPring-8 standard system in September 2000. As a result described in below. At this time, we also introduced the of this integration, the SPring-8 accelerator complex was standard database system for the linac control [3,4]. controlled by one unified system. Because the linac was Because the linac played a role as an injector to the continuously running as the electron beam injector not booster synchrotron and the NewSUBARU storage ring, it only for the SPring-8 storage ring but also for was necessary to avoid downtime due to the integration NewSUBARU, we had to minimize the hardware procedures. We separated the hardware upgrade work into modification to reduce the time for the development and two phases, that is, we replaced the system with minimum testing of the new control system. The integration method modification at the first phase and postponed an overall was almost the same as that of the integration of the replacement to the near future.
    [Show full text]
  • RIP Architecture L Technical Information
    L Technical Information RIP Architecture A raster image processor (RIP) is a computer with a very specific task to perform. It translates a series of computer commands into something that an output device can print. For an imagesetter, this means the instructions that instruct the laser to expose certain areas on the film. Acronyms ASIC Application Specific Integrated Circuit RIP architecture refers to the CISC Complex Instruction Set Computer 1An excellent source of design of the RIP, particularly CPU Central Processing Unit information on computer EPROM Erasable Programmable Read-Only technology is The Computer the hardware and software Memory Glossary by Alan Freedman. that make the RIP work. It is I/O Input/Output It is available from The impossible to discuss RIP MHz Megahertz Computer Language architecture without MIPS Million Instructions Per Second Company, (215) 297-5999. mentioning a number of 1 (also the name of a company that computer buzz words , many manufactures CPUs) of which are acronyms (see PROM Programmable Read-Only Memory list to right). And so to begin, RIP Raster Image Processor we’ll start with a review of RISC Reduced Instruction Set Computer computer basics, and see how SPARC Scalable Performance ARChitecture those principles apply to RIPs. TAXI Transparent Asynchronous Xmitter receiver Interface XMO eXtended Memory Option Computer basics The CPU of a computer is where the bulk of the computing gets done. Microprocessor designations The CPU may also be called the Intel Corporation microprocessor if the CPU is built on a • 8080, 8088 single miniaturized electronic circuit (i.e. • 8086, 80286, 80386, 80486 chip).
    [Show full text]
  • Porting the Choices Object-Oriented Operating System to the Motorola 68030
    PORTING THE CHOICES OBJECT-ORIENTED OPERATING SYSTEM TO THE MOTOROLA 68030 BY BJORN ANDREW HELGAAS B.A., Augustana College, 1988 THESIS Submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science in the Graduate College of the University of Illinois at Urbana-Champaign, 1991 Urbana, Illinois TABLE OF CONTENTS Chapter 1 Introduction .......................................... 1 2 Choices Overview ....................................... 4 2.1 Processes ........................................ 5 2.2 Exceptions ....................................... 7 2.3 Virtual Memory .................................... 8 3 Process Scheduling ...................................... 10 3.1 Choices Process Scheduling .............................. 12 3.2 The MC68030SystemContext Class .......................... 15 3.3 Summary ........................................ 17 4 Exception Handling ..................................... 19 4.1 Choices Exception Handling .............................. 20 4.2 MC68030 Exception Handling ............................. 23 4.3 The MC68030CPU Class ................................ 25 4.4 Summary ........................................ 31 5 Virtual Memory ....................................... 32 5.1 Choices Virtual Memory ................................ 34 5.2 MC68030 Virtual Memory ............................... 35 5.3 The MC68030Translation Class ............................ 39 5.4 The MC68030MMU Class ............................... 42 5.5 Summary .......................................
    [Show full text]
  • 32-Bit Bus Master Ethernet Interface for the 68030
    (Using the Macintosh SE/30)32-Bit Bus Master Ethernet Interface for the 68030 AN-691 National Semiconductor 32-Bit Bus Master Ethernet Application Note 691 Interface for the 68030 William Harmon (Using the Macintosh April 1993 SE/30) OVERVIEW National Semiconductor's SE/30 Ethernet adapter provides with any AUI interface. In fact, the SE/30 board has the a high performance, 32-bit, bus master network connection capability to be connected to a network through either thin for Apple's 68030 based compact Macintosh computer. wire (10Base2) or AUI drop cable (10Base5) Ethernet. This design is based around National Semiconductor's Sys- It is also worth mentioning that the SE/30 adapter supports tems Oriented Network Interface Controller (SONICTM, the use of Macintosh Nubus Slot Manager features, such as DP83932), which interfaces directly to the extension slot of interrupt handling, with an on board Slot Manager PROM. the SE/30. The SE/30 design also serves as a model for This does not cause the board to incur any extra cost, since designing the SONIC onto the mother board of a 68030 some type of PROM must already be used to store the based system, since the SE/30's one expansion slot is es- adapter's Ethernet address. sentially a direct connection to the Motorola 68030. FEATURES A block diagram of the adapter can be seen in Figure 1. The SE/30 Ethernet adapter operates synchronously with the Y 32-bit bus master system interface 16 MHz SE/30 mother board and accesses the necessary Y Asynchronous high speed 3 cycle DMA transmit and receive buffers directly in the system's main Y 100% on card address filtering, via the SONIC's on memory, via 16 MHz 3 cycle asynchronous DMA opera- board Content Addressable Memory (CAM) tions.
    [Show full text]
  • Energy Consumption of Apple Macintosh Computers
    Energy Consumption of Apple Macintosh Computers Jacob R. Lorch Alan Jay Smith Report No. UCB/CSD-97-961 June 1997 Computer Science Division (EECS) University of California Berkeley, California 94720 Energy Consumption of Apple Macintosh Computers Jacob R. Lorch and Alan Jay Smith Computer Science Division, EECS Department, University of California, Berkeley Berkeley, CA 94720-1776 June 17, 1997 Abstract easily visible without a backlight, a low-power proces- sor, a low-power disk, ¯ash memory instead of mechan- The utility of a portable computer is critically depen- ical disk, etc. Second, one can take better advantage dent on the period it can be used while running off the of the low-power states of existing system components. battery. In this paper, we present a study of power con- For example, most hard drives made for portable com- sumption in Apple Macintosh computers. We measure puters can be made to stop spinning in order to save the existing power consumption for each system com- power, and operating systems can take advantage of ponent using built-in measuring tools. Since total power this by spinning down the hard drive when it is not in consumption is a function of user workload, we use eight use [17]. Another low-power state that can be taken ad- user workload traces to determine power use as observed vantage of is running the CPU at a lower voltage. Power in practice. Apple currently implements some power- consumption drops with the square of the voltage, but saving features, and the effectiveness of those features the maximum clock rate also drops with the voltage.
    [Show full text]