Acceleration of system interrupts between operating systems in guest-host relationship

• US 5,038,281 A
• Filed: 09/19/1986
• Issued: 08/06/1991
• Est. Priority Date: 09/19/1986
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Peters, Anthony M.
• Primary Examiner(s): Eng, David Y.
• Application Number:
  US06/909,523
• Time in Patent Office: 1,782 Days
• Field of Search: 364/200 MS File, 364/900 MS File
• US Class Current: 710/260
• CPC Class Codes: G06F 9/45537 Provision of facilities of other operating environments, e.g. WINE I/O emulation G06F13/105
• CPC Class Codes: G06F 9/4812 by interrupt, e.g. masked

Abstract

This disclosure describes the acceleration of system interrupts between one operating system and another operating system which run in a guest-host relationship. The execution time required by the host operating system (HOS) to service system interrupts is substantially reduced, thus allowing the guest operating system (GOS) to execute more efficiently. The invention is implemented by enhancing HOS supervisor services and HOS dispatching functions so that GOS resident supervisor functions are bypassed.

Claims

1. In a data processing system having a guest operating system (GOS) running under a host operating system (HOS), said GOS including a user level and a system nucleus in which system interrupts are used to service requests between the user level and the nucleus, a method for accelerating system interrupts between said GOS and HOS comprising the steps of:

identifying in the HOS each GOS interrupt issued as a result of a user-to-nucleus request; and

then dispatching the request directly from the HOS to the GOS nucleus.

2. A method as in claim 1 further comprising:

identifying in the HOS each interrupt issued as a result of a nucleus-to-user request; and

dispatching the request from the HOS directly to the GOS user.

3. A method as in claim 1 further comprising, prior to said dispatching:

accessing, by the HOS, information contained in the GOS relating to timing; and

performing timing calculations required by the GOS in the HOS.

4. A method as in claim 1 wherein said GOS stores a program status work indicating an interrupt type, said method further comprising, prior to said dispatching:

converting the program status work stored by the GOS into a new program status work in the HOS.

5. In a data processing system having a guest operating system (GOS) running under a host operating system (HOS), said GOS including a user level and a system nucleus of a kernel and a resident supervisor in which interrupts are used to service requests between the user level and the nucleus, a method for accelerating system interrupts between said GOS and HOS comprising the steps of:

identifying the HOS each GOS interrupt issued as a result of a user-to-kernel request; and

then dispatching the request directly from the HOS to the GOS kernel, bypassing the resident supervisor.

6. A method as in claim 5 further comprising;

identifying the HOS each interrupt issued as a result of a kernel-to-user request; and

dispatching the request from the HOS directly to the GOS user, bypassing the resident supervisor.

7. A method as in claim 5, further comprising, prior to said dispatching:

accessing, by the HOS, information contained in the GOS relating to timing; and

performing timing calculations required by the GOS in the HOS.

8. A method as in claim 5 wherein said GOS stores a program status word indicating an interrupt type, said method further comprising, prior to said dispatching:

converting the program status work stored by the GOS into a new program status word in the HOS.