System and method for supporting real-time computing within general purpose operating systems

• US 6,466,962 B2
• Filed: 06/07/1995
• Issued: 10/15/2002
• Est. Priority Date: 06/07/1995
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bollella, Gregory
• Primary Examiner(s): Follansbee, John A.
• Application Number:
  US08/472,365
• Publication Number:
  US 20010054055A1
• Time in Patent Office: 2,687 Days
• Field of Search: 395/800, 395/650, 395/500, 395/670, 370/58.2, 709/226, 709/102, 709/104, 709/107, 709/108
• US Class Current: 718/107
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/4825 Interrupt from clock, e.g. time of day
• CPC Class Codes: G06F 9/4843 by program, e.g. task dispatcher, supervisor, operating system
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455

Abstract

A method of supporting real-time computing within a general purpose operating system, by supporting co-resident operating systems. This is a methodology for allowing existing, well-understood real-time systems technology to co-exist with commercial, general purpose operating systems to support applications such as desktop multimedia conferencing. The approach is to partition the central processor and other system resources into two virtual machines, first a machine running a largely unmodified general purpose operating system, and second a machine running a real-time kernel, then to multiplex access to the physical hardware by the virtual machines.

Claims

1. A method of providing real-time computing support within a data processing system, wherein said data processing system includes a processor, a first operating system, a second operating system a real-time clock, and a plurality of tasks, wherein said first operating system is a general purpose operating system which operates on a non-real-time basis, and said second operating system operates on a real-time basis, said method comprising the steps of:

periodically determining real-time computing support requirements of each of said plurality of tasks within said data processing system utilizing a resource allocation model; and

dynamically scheduling and allocating system resources within said data processing system to each of said first operating system and said second operating system on a regular and precise basis as determined by said real-time clock, in response to said determination step, such that said first operating system has control of said processor for a predetermined and selected portion of a cycle of operation and said second operating system has control of said processor for a remainder portion of said cycle of operation.

2. A method according to claim 1 wherein said first operating system has at least a first task executing thereunder, and said second operating system has at least a second task executing thereunder.

3. A method according to claim 2 wherein said first task and said second task share a memory space.

4. A method according to claim 3 including the step of checking a memory reference to said memory space while executing said second task to make sure that said memory reference does not conflict with use of said shared memory space by said first task.

5. A method according to claim 1, said method further comprising the steps of:

generating a repetitive time marker having a first period; and

performing said determination step and said allocation step in response to said repetitive time marker.

6. A data processing system having a processor, a first operating system, a second operating system, a real time clock, and a plurality of tasks, wherein said first operating system is a general purpose operating system, which operates on a non-real-time basis, and said second operating system operates on a real-time basis, said data processing system comprising:

means for periodically determining real-time computing support requirements for each of said plurality of tasks within said data processing system utilizing a resource allocation model; and

means for dynamically scheduling and allocating system resources within said data processing system to each of said first operating system and said second operating system on a regular and precise basis as determined by said real-time clock, in response to said determination step, such that said first operating system has control of said processor for a predetermined and selected portion of a cycle of operating and said second operating system has control of said processor for a remainder portion of said cycle of operation.

7. A system according to claim 6 wherein said first operating system has at least a first task executing thereunder, and said second operating system has at least a second task executing thereunder.

8. A system according to claim 7 wherein said first task and said second task share a memory space of said memory.

9. A system according to claim 8 including means for checking a memory reference to said memory space while executing said second task to make sure that said memory reference does not conflict with use of said shared memory space by said first task.

10. A system according to claim 6, said system further comprising:

means for generating a repetitive time marker having a first period; and

means for performing said determination means and said allocation means in response to said repetitive time marker.