首页> 外国专利> RESERVATION OVERRIDING NORMAL PRIORITIZATION OF MICROPROCESSORS IN MULTIPROCESSING COMPUTER SYSTEMS

RESERVATION OVERRIDING NORMAL PRIORITIZATION OF MICROPROCESSORS IN MULTIPROCESSING COMPUTER SYSTEMS

机译:多处理计算机系统中微处理程序的预留优先级高于常规优先级

摘要

the invention uses three programs of priority to determine which of several processing units receives the priority to become the master processor bus of a host bus in a multiprocessor system, and program for transferring control arbitration a master of the bus to another.each ranking priority class of elements, where a total of (n / 2) x (n + 1) bits of the priority controls the relative priority between each pair of elements. a component receives the highest priority in each of the n-1 bits of the priority associated with that element is.in the bus arbitration, teacher strength determines the moment at which the transfer of control of the host bus is such that it is governed by one of the programs are ranked in order of priority.the program provides for arbitration control bus eisa, regeneration of ram memory and direct memory access dma greater priority for units serving as bus masters and a bus master can interrupt the temporary bus master in neck so for a rewrite cache intervention cycle.the program also provides an arbitration processing pipeline of addresses, a transmission burst, split transactions aborted and cpu reservations, in the test of a cycle lock.the processing pipeline allows the next address bus to address control signal and the state before the start of the next phase of data transfer from the bus master.split transactions are applied to a central unit, a reading on the bus is assigned after the arbitration bus, host to another device without the need for a ru00e9arbitrage host bus to retrieve the data.the data is transmitted on the bus when it is inactive, the host, even when the bus is controlled by a host device.
机译:本发明使用三个优先级程序来确定几个处理单元中的哪个接收到优先级,以成为多处理器系统中主机总线的主处理器总线,以及用于将控制仲裁的主机转移到另一个的程序。元素,其中总共(n / 2)x(n + 1)位优先级控制每对元素之间的相对优先级。在与该元素关联的优先级的n-1位中的每一个中,组件都接收到最高优先级。在总线仲裁中,老师的力量决定了主机总线的控制权转移受其控制的时刻。其中一个程序是按优先级排列的。该程序提供仲裁控制总线eisa,ram内存的再生和直接内存访问dma的较高优先级,用作充当总线主控器的单元,并且总线主控器可以中断临时的总线主控器,因此在重写缓存干预周期中,该程序还提供了一个地址,传输突发,拆分事务中止和cpu保留的仲裁处理流水线,以进行周期锁定测试。该处理流水线允许下一个地址总线寻址控制信号和从总线主控器进行下一阶段数据传输之前的状态。将拆分事务应用于中央单元,在总线后部分配读数在仲裁总线上,主机无需使用仲裁套用主机总线即可检索到另一台设备的数据。当数据处于非活动状态时,即使在总线由主机设备控制的情况下,数据也会在主机上传输。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号