Home | Links | Contact Us | More About Intellectual Property | Bookmark
Search patents:
Home I/O Systems Arbitration-circuitry-for-deciding-access-requests-from-a-multiplicity-of-components

 Binary MOS ripple-carry parallel adder/subtracter and adder/subtracter stage suitable therefor
What is claimed: 1. A parallel adder/subtracter employing enhancement-mode insulated-gate field-...


 System for controlling power distribution to customer loads
The foregoing and other objects of the invention are attained in accordance with one aspect of the ...


 Apparatus for controlling the time sequenced energization of a memory unit
It is, accordingly, an object of the present invention, to provide an improved sequencing control ...


 Active selectable digital delay circuit
What is claimed is: 1. A multiplexer for an active selectable digital delay circuit comprising: a ...


 Method for selectively saving/restoring first registers and bypassing second registers in register units based on individual lock/unlock status thereof
OF THE PREFERRED EMBODIMENTS Referring now to the drawings wherein the showings are for purposes ...


 Computer system capable of connecting expansion unit
It is an object of the present invention to provide a computer system capable of connecting an ...


 Device and method for providing a simulation of an idle UART to prevent computer lockup
OF A PREFERRED EMBODIMENT The present invention prevents an operating system such as Windows.RTM. 9...


 Low-power consuming information processing apparatus
It is an object of the first embodiment to solve the above-described conventional problems and ...


 Power management process
The underlying concept of the invention is to subtract the torque (or power) required to drive the ...


 Method and apparatus for selecting nodes in configuring massively parallel systems
To address the requirements described above, the present invention discloses a simplified method, ...


 Arbitration circuitry for deciding access requests from a multiplicity of components

Details
Inventors: Capizzi, Giuseppe N.; Melgara, Marcello;
Assignee: CSELT Centro Studi E Laboratori Telecomunicazioni S.p.A. (Turin, IT)
Primary Examiner: Eng; David Y.
Assistant Examiner:
Attorney, Agent or Firm: Ross; Karl F., Dubno; Herbert

A circuit arrangement for deciding concurrent requests for access to a common data bus emitted by a number n=2.sup.m of components of different ranks, specifically multiprocessor elements, comprises n mutually identical arbitration devices respectively associated with these components. The arbitration devices are interconnected in a chain by a priority bus of branched binary structure divided into m lines connected, within each device, to a logic network that is also connected to an internal m-conductor bus extending from a priority-code register. The priority bus, whose lines are normally at zero potential, is connected in any given device to the register thereof in the presence of an access request from the associated component whereupon its logic network determines whether the code on that bus equals the contents of the register; if so, the associated component is enabled by a control unit of the device to access the data bus. If, however, a simultaneous access request from a higher-ranking component causes another arbitration device to energize the priority bus, the code emitted by that other device overrides that of the first-mentioned device whose logic network therefore detects an inequality. A cascaded connection of the control units of the several arbitration devices enables a decrementer in each device to establish priority codes of progressively lower rank along the cascade in an initialization phase; the logic network also modifies, at the beginning of each operating cycle except during prolonged seizure of the data bus by any component, the priority code initially assigned to each arbitration device in order to give precedence to nominally lower-ranking components.

DETAILED DESCRIPTION We claim: 1.
In a data-handling system with a multiplicity n=2.
sup.
m of components of different ranks served by a common data bus alternatively accessible by said components in an order of precedence based on their ranks, the combination therewith of a chain of n mutually identical arbitration devices respectively associated with and individually connected to said components, said arbitration devices being interlinked by a priority bus with m lines having a binary-tree structure including a first line connected to all n devices and an m.
sup.
th line divided into n/2 sections each connected to a pair of adjacent devices, said devices following one another in a predetermined ranking order of the associated components, each of said arbitration devices comprising: register means for storing an m-bit priority code initially assigned to the respective device, the priority codes initially assigned to consecutive devices having numerical values varying progressively in conformity with said ranking order of the associated components; normally blocked gating means inserted between said priority bus and an output circuit of said register means for transmitting the stored priority code to said priority bus in the presence of an access request from the associated component; comparison means connected to said priority bus and to said output circuit for detecting an equality between the transmitted priority code and the state of energization of the lines of said priority bus in the presence of said access request, an arbitration device associated with a higher-ranking component modifying said state of energization by an overriding priority code simultaneously transmitted to the priority bus; and a control unit responsive to said access request for unblocking said gating means and sending a consent signal to the associated component in the presence of an equality determination by said comparison means, said consent signal enabling the associated component to access said data bus, each arbitration device further comprising calculating means having reference inputs connected to said priority bus inserted in series with a normally inactive driver in a loop between said output circuit and an input circuit of said register means for reloading same upon activation of said driver by a command from said control unit, said calculating means implementing the Boolean equation ##EQU1## where * stands for the AND function, XOR stands for the Exclusive-OR function, p



Related patents
  Paged memory management unit which locks translators in translation cache if lock specified in translation table
Accordingly, it is an object of the present invention to provide a mechanism which allows a paged memory management unit to determine automatically from a field in each ...
  Ensuring data integrity by locked-load and conditional-store operations in a multiprocessor system
In accordance with one embodiment of the invention, a high-performance processor is provided which is of the RISC type, using a standardized, fixed instruction size, and ...
  Multiplexing communication card and scanning method for run-in testing
i The present invention provides a multiplexing communication card and scanning system for testing and monitoring the test results of many PC's using a single host ...
  Fixture for motor controller power substrate and motor controller incorporating
In accordance with a first aspect of the invention, a fixture is provided for securing conducting pins to a power substrate module of the type including a rectifying ...
  Asynchronous digital time-division multiplexing system with distributed bus
What we claim is: 1. An asynchronous digital time-division multiplexing system, comprising (a) plural data sources each having active and inactive states and being ...
  Method for executing overlays in an expanded memory data processing system
OF THE INVENTION FIG. 1 shows an Expanded Memory System (EMS). The system includes a one megabyte (1024K) physical memory 1 address space and a typically larger, for ...
  Robot program checking method
An object of the present invention is to provide a robot program checking method which permits accurate and rapid edition of a robot program. In order to achieve the ...
  Method of correcting machine position change
The present invention has been made in view of the aforesaid drawbacks, and an object thereof is to provide a method of correcting a machine position change in which an ...
  Communications controller utilizing an external buffer memory with plural channels between a host and network interface operating independently for transferring packets between protocol layers
OF ILLUSTRATIVE EMBODIMENT Referring to FIG. 2, the programmable (VLSI) data communication controller chip 15 of the present invention, hereinafter "data communication ...
  Fault diagnosis apparatus and method for sequence control system
Accordingly, it is an object of the present invention to eliminate the disadvantages of the conventional methods/apparatuses and to shorten the time required for fault ...

0.014

Archive: All patents - Links

Copyright (c)2006 Eipa-patents.org - All rights reserved