2001年微处理器展望

对２００１年微处理器进行展望。如果６４位Ｘ８６ＩＡ－６４）上市，由ｘ８６同ＲＩＳＣ性能将相关无几，如果ＷｉｎｄｏｗｓＮＴ和Ｉｎｔｅｒｎｅｄ机普及，则处理器将比现有有更多选择。     

SGS－THOMSON公司宣布5X86和6X86第二来源计划

ＳＧＳ－ＴＨＯＭＳＯＮ公司宣布５Ｘ８６和６Ｘ８６第二来源计划自从ＳＴ４８６ＤＸ系列成功进入ｘ８６市场以后后；ＳＧＳ－ＴＨＯＭＳＯＮ公司宣布了针对下一代处理器的第二来源计划。新一代的处理器，即ＳＴ５ｘ８６和ＳＴ６ｘ８６，将分别于１９９６年第一季度和第二...     

要闻简报

●Ｃｙｒｉｘ公司日前推出６ｘ８６微处理器新品系列６ｘ８６－Ｐ１３３、６ｘ８６－Ｐ１５５及６ｘ８６－Ｐ１６６、据著名分析公司ＭｉｃｒｏＤｅｓｉｇｎ　Ｒｅｓｏｕｒｃｅｓ实验室评测，称６ｘ８６较Ｐｅｎｔｉｕｍ更为出色：●美国著名的微处理器芯片生产供应商Ｉｎｔｅｌ公司将在上海浦东外高桥保税区南部区域建立集成电路生产和销售服务基地。目前，该公司已与上海市外高桥保税区新发展有限公司正式签订合同。这是继ＨＰ公司之后进驻保税区的又一家美商独资的高科技企业。（夏红）●ＴＩ公司宣布，去年业绩达创新高，净营业额为１３…     

Cyrix 6x86处理器确立新的台式PC机性能标准——测试结果表明,Cyrix 6x86处理器的性能在每个类别中都明显优于Pentium处理器

Ｃｙｒｉｘ公司的高性能６ｘ８６处理器家族又增添了３个新成员：６ｘ８６－Ｐ１３３＋、６ｘ８６－Ｐ１５０＋和６ｘ８６－Ｐ１６６＋。６ｘ８６－Ｐ１５０＋和６ｘ８６＝Ｐ１６６＋是第一批由主要的工业分析公司ＭｉｃｒｏＤｅ－ｓｉｇｎ　Ｒｅｓｏｕｒｃｅｓ的ＭＤＲ实验室使用新的Ｐ－ｒａｔｉｎｇ（Ｐ评级）系统独立评估的处理器。测试结果稳固地确立了６ｘ８６处理器性能超越Ｐｅｎｔｉｕｍ处理器的领先地位。除了ＭＤＲ实验室公布的６ｘ８６Ｗｉｎｓｔｏｎｅ９６结果外，Ｃｙｒｉｘ还宣布了充分突出新型６ｘ８６处理器的３２位性能的…     

SGS-THOMSON公司支持6x86处理器成为Pentium处理器的替代产品

ＳＧＳ－ＴＨＯＭＳＯＮ　Ｍｉｃｒｏｅｌｅｃｔｒｏｎ－ｉｃｓ公司全面支持６ｘ８６体系结构作为Ｐｅｎｔｉｕｍ的替代结构，并深信很多ＰＣ机、工作站和服务器制造厂商和其他ＯＥＭ厂商将发现ＳＴ６ｘ８６是满足其系统需求的切实可行的解决方案。ＳＧＳ－ＴＨＯＭＳＯＮ对ＭＤＲ实验室完成的独立基准程序测试结果感到非常高兴。这些测试证实，在相同的系统配置中，６ｘ８６在每个器件类别中的性能都高于Ｐｅｎｔｉｕｍ。为了使客户能方便地对照比较不同ｘ８６体系结构的性能，ＳＧＳ－ＴＨＯＭ－ＳＯＮ、Ｃｙｒｉｘ、ＩＢＭ和ＡＭＤ已联合…     

Intel x86CPU系列发展历程

介绍了以Ｉｎｔｅｌ ｘ８６ＣＰＵ２０多年的发展历程，对各代（ｘ８６）从多角度进行了分析比较，阐明其性能的提高以及的采用的新技术。使读者了解到，在这２０多年里，周密的设计技术使ＰＣ机性能有了很大的提高。     

我电路交换技术获重大突破

深圳市中兴通讯股份有限公司南京研究所近日成功研制开发出６４Ｋｘ６４Ｋ交换网板，这是我国电路交换技术的一项重大突破。此前，业界已投入应用的容量最大的交换网板为１６Ｋ×１６Ｋ。由于单Ｔ结构的交换网络比多级交换网络具有许多不可比拟的优点，因此设计出一种结构简单的单Ｔ大容量交换网一直是该领域追求的目标。但是，大客量的单Ｔ交换网板技术实现难度大。中兴通讯集多年交换机硬件开发经验，大胆转变传统设计思路采用独创的新技术简化设计，成功地在一块普通单板上完成了６４Ｋｘ６４Ｋ交换容量，并且全部采用通用器件。该交换网…     

微机与多8031单片机系统中的C＋＋与80x86汇编语言互连

结合开发电视台节目制作中的电源控制实时系统工作，说明如何在微机与多８０３１实时控制系统中使Ｃ＋＋与８０ｘ８６汇编语言实现互连，并且给出了软件实现。     

Oracle与xBASE的数据转换

Ｏｒａｃｌｅ与ｘＢＡＳＥ的数据转换●王斌如何将现有的ｘＢＡＳＥ的数据库转换到Ｏｒａｃｌｅ数据库中，这是一个很现实的问题。要解决这一问题，如果是在Ｗｉｎｄｏｗｓ下的Ｆｏｘｐｒｏ软件，可以使用ＯＤＢＣ的方式实现ｘＢＡＳＥ到Ｏｒａｃｌｅ的转换，但是如果在Ｕ...     

高升公司推出信息龙处理器

近日 ,美国高升公司推出了与ｘ８６兼容的信息龙(ｉＤｒａｇｏｎ)处理器 ,这是目前耗电量最低的信息家电和多媒体ｘ８６处理器 ,也是第一个可用一节１．５Ｖ电池驱动的ｘ８６多媒体处理器。这一芯片设计拥有数十个专利 ,可应用到不能有风扇的多媒体信息家电产品和用户终端上。其ＣＰＵ核具备了很强的Ｊａｖａ编程能力、多媒体和因特网浏览能力、高效的ＭＰＥＧ４视频数据流处理能力以及大幅度提高系统可靠性的热管理能力。高升公司已开发出系统集成芯片 (ＳＯＣ)系列及高画质ＭＰＥＧ４软件技术 ,可用于用户终端视频点播(ＶＯＤ)、高清晰度…     

A Machine State Transition Approach to Instruction Retargeting for Embedded Microprocessors

This paper presents an interesting approach to retargeting existing software at the assembly (or binary) level from one instruction set to another instruction set. The approach is based on abstracting the instruction set behaviors as symbolic transitions of the machine states. The retargeting process is modeled as a planning process, an AI technique, that finds a plan (a sequence of operations) which brings the target processor from the same initial state to the same final state as the original software does on the source processor. The approach has been successfully applied in a design project of an x86 compatible microprocessor with an embedded internal RISC core for efficient execution. The proposed approach produced optimal x86-to-RISC mapping. In addition, the approach made it easy to keep up with microarchitecture revision during the design exploration phase since the mapping table can be automatically re-generated and re-evaluated promptly, which is difficult to achieve manually.     

面向基于x86处理器和AMBA的系统芯片的全系统模拟器PKUsim-86

 基于周期级全系统模拟器对微体系结构进行系统性能评估成为芯片设计必不可少的环节.虽然x86处理器是当前商业和科学计算领域最广泛采用的处理器,很少有开源的x86模拟器能够满足研究需要.本文面向基于Geode GX x86处理器和AMBA总线的PKUnity-86系统芯片,设计并实现了周期级全系统模拟器PKUsim-86.它可以启动Microsoft DOS、Windows 98、Windows XP等操作系统,运行典型的x86应用程序.PKUsim-86支持功能模拟和性能模拟的在线切换,其指令模拟速度为0.86MIPS,与真实硬件的对比表明,PKUsim-86具有较高的相对准确度.     

Chill Out

Heat is one of the worst enemies of electronics. Sit on the sofa with your laptop and you quickly feel the heat on your lap. Often though, overheating can be hard to diagnose. You may notice random errors occurring no matter what program you're running. This is especially true if you use your computer to play advanced video games, which can really tax the microprocessor and the graphics card. If your machine frequently experiences fatal errors or "the blue screen of death" on such occasions, chances are it has thermal management problems. And overheating doesn't just degrade a computer's performance; it can also shorten its useful life.     

Multi-processor systems

Microprocessor software is becoming more complex as the one unit is used to handle a multiple of tasks. This leads to considerable overhead in task scheduling software. In addition, communication between separate microprocessor systems also requires extra software.If, however, two or more microprocessors may be attached to the same bus, then many disadvantages of single microprocessor realisations are eliminated.These points, together with practical applications for systems and peripheral controllers are discussed     

电子式微处理器热继电器

应用微处理器微电子技术制作热继电器,克服双金属片热继电器的缺陷。分析电机发热的主要原因,通过微处理器和电子电路对电机进行保护。     

基于嵌入式Linux的无线检测系统设计

针对非接触测量场合和智能检测的需要,设计了基于ARM处理器和嵌入式Linux的无线数据采集系统。该系统采用Atmel AT91RM9200微处理器,并在Linux环境下实现网络通信和数据检测功能,并通过ARM处理器与智能RF芯片AT86RF211的接口设计完成数据的无线检测。详细介绍了ARM9微处理器与AVR MCU,智能RF芯片,以太网接口的软硬件设计以及在Linux下的整个智能网络监测系统的实现过程。     

On Applying Graph Theory to ILP Analysis

The evaluation of computer architectures requires new tools that complement the customary simulations. Graph theory can help to create a new frame of fine grain parallelism analysis. The differences found between the superscalar performance in x86 and non-x86 processors and the peculiar characteristics of the x86 instruction set architecture recommend to carry out a thorough study of the available parallelism at the machine language layer. Starting off from graph theory foundations, new concepts are introduced, from reduced valence to data dependence matrix D, the latter characterizing a code sequence in a mathematical manner. This matrix satisfies a series of properties and restrictions and provides information about the ability of the code to be processed concurrently. The different sources of data dependencies can be composed, facilitating a way to analyze their final influence on the degree of parallelism.     

Technique for performing multiplication on a 16-bit microprocessor using an extension of Booth's algorithm

Multiplication of a 2's-complement number by a known coefficient is often required in digital signal processing and other applications. When this is required to be performed on a microprocessor, an extension of Booth's algorithm (e.b.a.) can be used. Examples are presented to demonstrate how the multiplication is performed using e.b.a. on a microprocessor, and comparisons are made with the machine multiply instruction (m.m.i.) which show that the e.b.a. can take significantly less machine time.     

6×86: the Cyrix solution to executing ×86 binaries on ahigh performance microprocessor

With the 6×86 microprocessor, Cyrix's design team demands optimum functionality and performance at an acceptable cost. Cyrix maintains that the way to get high performance is to keep larger units of processing together and to incorporate as much concurrency of execution as possible. To do the former, the microprocessor attempts to keep together all parts of an ×86 instruction as it passes through its seven processing steps. To do the latter, it attempts to initiate the processing of two ×86 instructions each cycle, and pipelines the processing of each instruction in seven stages. Most of this paper deals with issues involving keeping these seven stages of the two pipelines busy so the microprocessor can attempt to approach the twofold increase in performance that is made possible by the underlying structure     

The design of the microarchitecture of UltraSPARC-I

The realization of a high performance modern microprocessor involves hundreds of person-years of conception, logic design, circuit design, layout drawing, etc. In order to leverage effectively the 5-10 millions of transistors available, careful microarchitecture tradeoff analysis must be performed. This paper describes not only the microarchitecture of UltraSPARC-I, a 167 MHz 64-b four-way superscalar processor, but more importantly it presents the analysis and tradeoffs that were made “en route” to the final chip. Among several issues, the in-order execution model is compared with alternatives, variations of the issue-width of the machine as well as the number of functional units are described, subtle features that are part of the memory hierarchy are explained, and the advantages of the packet-switched interconnect are exposed