Time multiplexed processor bus

• US 4,630,193 A
• Filed: 09/23/1982
• Issued: 12/16/1986
• Est. Priority Date: 04/27/1981
• Status: Expired due to Fees

• Current Assignee: Textron Inc
• Original Assignee: Textron Inc
• Inventors: Kris, Bryan
• Primary Examiner(s): Heckler, Thomas M.
• Application Number:
  US06/428,488
• Time in Patent Office: 1,545 Days
• Field of Search: 364/200 MS File, 364/900 MS File
• US Class Current: 713/502
• CPC Class Codes: G06F 13/26 with priority control
• CPC Class Codes: G06F 13/372 using a time-dependent priority, e.g. individually loaded time counters or time slot

Abstract

A multi-master processor bus and a method of processing data which permits multiple microprocessors to communicate freely and inexpensively among themselves and various system resources. The bus uses a multiphase clock and latches to provide time slice signals to sequentially activate each processor, one at a time in a repetitive sequence. The bus includes cables and terminals for each of the cables with means for interconnecting each of the modules in series daisy chain fashion to selected cables.

Claims

1. A synchronous data processing system comprising:

N processor modules, each processor module including a processor and interface circuitry connected with the processor;

at least one resource module available for access by the processor modules;

a signal global microprocessor bus system commonly shared by the resource module and each of the processor modules, the resource module and the interface circuitry of each of the processor modules being connected in parallel to the bus system to enable communication among the processor modules and between the processor modules and the resource module; and

system timing means connected with the bus system for dividing a predetermined bus cycle time period into N equal time slices, each processor module being granted access to the global bus for a single time slice during each bus cycle time period;

the interface circuitry of each processor module being configured to permit each processor to execute data received from a resource module directly as an instruction and being configured to enable each processor module to communicate information in signle word messages during the processor module'"'"'s respective time slice so that the length of communication among any two of the processor modules and between any processor module and the resource module does not inhibit access by any other processor module to the resource module and to either of said two processor modules.

2. The data processing system as recited in claim 1 wherein the processor modules are each assigned a module number running sequentially from 0 to N-1, and the time slices are each assigned a number running sequentially from 0 to N-1, each processor module being granted access to the global bus during the numbered time slice which corresponds to its module number.

3. The data processing system as recited in claim 2 wherein:

the system timing means includes a counter for generating consecutive repeating count signals from 0 to N-1, the counter outputting the count signals onto the global bus system; and

decoder circuitry within the interface circuitry of each processor module for receiving the count signals and for granting access to the global bus system when a count signal corresponds to the module number of the particular processor.

4. The data processing system as recited in claim 2 wherein the predetermined bus cycle time period is 500 nanoseconds.

5. The data processing system as recited in claims 1 or 4 wherein the global bus system comprises a time slice and clock bus, a data bus and address bus, each of said buses including a plurality of lines.

6. The data processing system as recited in claim 2 wherein the processor modules are all of the same structure and wherein the module numbers are assigned in a completely arbitrary manner.

7. The data processing system as recited in claim 1 wherein each processor module is fully self-sufficient and capable of asynchronous operation itself without regard to the global bus.

8. A synchronous data processing system comprising:

N processor modules, each processor module including a processor and interface circuitry connected with the processor;

a single global microprocessor bus system commonly shared by each of the processor modules, the interface circuitry of each of the processor modules being connected in parallel to the bus system to enable communication among the processor modules;

system timing means connected with the bus system for dividing a predetermined bus cycle time period into N equal time slices, each processor module being granted to access to the global bus for a single time slice during each bus cycle time period;

the interface circuitry of each processor module being configured to enable each processor module to communicate information in single word messages during the processor module'"'"'s respective time slice so that the length of communication between any two processor modules does not inhibit access by any other processor module to either of said two processor modules.