Methods and apparatus for distributing interrupts

• US 7,328,294 B2
• Filed: 12/03/2001
• Issued: 02/05/2008
• Est. Priority Date: 12/03/2001
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Kim, Ted, Gentry, Denton E. Jr.
• Primary Examiner(s): An; Meng-Al T.
• Assistant Examiner(s): Truong; Camquy
• Application Number:
  US10/011,899
• Publication Number:
  US 20030105798A1
• Time in Patent Office: 2,255 Days
• Field of Search: 395/733, 718/103, 718/104, 718/105, 710/206, 710/260, 710/268, 710/264
• US Class Current: 710/260
• CPC Class Codes: G06F 9/4812 by interrupt, e.g. masked
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals

Abstract

The present invention relates to handling interrupts in a multiprocessor system. An interrupt controller can receive input from a variety of interrupt sources, such as peripheral components and peripheral interfaces. Interrupts and their associated characteristics are identified. In one example, interrupt characteristics can be compared with characteristics of other interrupts handled by processors in the multiprocessor system. Interrupt characteristics are used to select a processor to run a routine for handling the associated interrupt. Intelligent selection provides efficient and effective distribution of interrupts.

Claims

1. A method for distributing interrupt load to processors in a multiprocessor system, the method comprising:

identifying a first characteristic associated with a first interrupt, wherein the first characteristic indicates that the first interrupt corresponds to an operation performed with a first device;

identifying a second characteristic associated with a second interrupt, wherein the second characteristic indicates that the second interrupt corresponds to an operation performed with the first device;

selecting a first processor for handling the first interrupt and a second processor for handling the second interrupt, wherein different processors are selected for handling interrupts from the same device to distribute load for interrupts from the same device, wherein the first characteristic identifies a scheduling group of the first interrupt and the second characteristic identifies a scheduling group of the second interrupt, wherein the first processor is selected based on whether the first processor is handling an interrupt having a scheduling group that is different from a scheduling group of the first interrupt, and wherein the second processor is selected based on whether the second processor is handling an interrupt having a scheduling group that is different from a scheduling group of the second interrupt; and

wherein the first and second characteristics also identify a priority of the interrupt, an originating peripheral causing the interrupt, a type of a corresponding interrupt handler routine, the scheduling group, or whether the corresponding interrupt handler routine was recently run.

2. The method of claim 1, wherein the second characteristic is maintained by an interrupt controller.

3. The method of claim 1, wherein the second characteristic is maintained by a communication adapter.

4. The method of claim 1, wherein the first characteristic comprises a protocol of the message triggering the first interrupt.

5. The method of claim 1, wherein the first characteristic comprises a session of the message triggering the first interrupt.

6. The method of claim 1, wherein the first characteristic comprises a priority of the first interrupt, wherein first processor is selected based on both the scheduling group and the priority of the first interrupt.

7. The method of claim 1, wherein the first processor is selected for handling the first interrupt if the first processor is handling an interrupt having a scheduling group that is different from the scheduling group of the first interrupt.

8. The method of claim 1, wherein comparing the first characteristic with the second characteristic comprises determining whether the characteristics are different.

9. The method of claim 1, wherein the first processor is configured to run an interrupt handler in parallel with a second processor, and wherein the first processor is selected based on both the scheduling group of the first interrupt and on whether the first processor has a warm cache with respect to an interrupt handler corresponding to the first interrupt.

10. An apparatus for distributing interrupt load to processors in a multiprocessor system, the apparatus comprising:

means for assigning a first characteristic to a first interrupt;

means for assigning a second characteristic to a second interrupt;

means for identifying the first characteristic associated with the first interrupt, wherein the first characteristic indicates that the first interrupt corresponds to an operation performed with a first device;

means for identifying the second characteristic associated with the second interrupt, wherein the second characteristic indicates that the second interrupt corresponds to an operation performed with the first device;

means for selecting a first processor for handling the first interrupt and a second processor for handling the second interrupt, wherein different processors are selected for handling interrupts from the same device to distribute load for interrupts from the same device, wherein the first characteristic identifies a scheduling group of the first interrupt and the second characteristic identifies a scheduling group of the second interrupt, wherein the first processor is selected based on whether the first processor is handling an interrupt having a scheduling group that is different from a scheduling group of the first interrupt, and wherein the second processor is selected based on whether the second processor is handling an interrupt having a scheduling group that is different from a scheduling group of the second interrupt; and

wherein the first and second characteristics also identify a priority of the interrupt, an originating peripheral causing the interrupt, a type of a corresponding interrupt handler routine, the scheduling group, or whether the corresponding interrupt handler routine was recently run.

11. The apparatus of claim 10, wherein if the first processor is handling an interrupt having a scheduling group that is different from the scheduling group of the first interrupt, then the means for selecting selects the first processor for handling the first interrupt.

12. The apparatus of claim 10, wherein the means for selecting selects the first processor based on both the scheduling group and the priority of the first interrupt.

13. The apparatus of claim 10, wherein the first processor is configured to run an interrupt handler in parallel with a second processor and wherein means for selecting selects the first processor based on both the scheduling group of the first interrupt and on whether the first processor has a warm cache with respect to an interrupt handler corresponding to the first interrupt.

14. The apparatus of claim 10, wherein comparing the first characteristic with the second characteristic comprises determining whether the characteristics are different.

15. The apparatus of claim 10, wherein the apparatus is an interrupt controller.

16. A method for dynamic selection of a processor in a multiprocessor system, the method comprising:

identifying a first interrupt from a communications adapter, the first interrupt including information identifying a communications session associated with the first interrupt;

identifying a routine for handling the first interrupt;

dispatching the first interrupt to a first processor upon checking a recent invocation indicator associated with the first processor to confirm that the first processor has a warm cache with respect to an interrupt handler corresponding to the first interrupt;

identifying a second interrupt; and

dispatching the second interrupt to the first processor if a first characteristic corresponding to the first interrupt and a second characteristic corresponding to the second interrupt are different and if the first and second interrupts are associated with different scheduling groups;

wherein the first and second characteristics identify a priority of the interrupt, an originating peripheral causing the interrupt, a type of a corresponding interrupt handler routine, the scheduling group, or whether the corresponding interrupt handler routine was recently run.

17. The method of claim 16, wherein the second interrupt is dispatched to the first processor if the first and second interrupts correspond to messages of different protocols.

18. The method of claim 16, wherein the second interrupt is dispatched to the first processor if the first and second interrupts are associated with different sessions.

19. The method of claim 16, wherein the second interrupt is dispatched to the first processor if the first and second interrupts have different priorities.

20. The method of claim 16, wherein the first processor is configured to running the routine for handling the interrupt in parallel with a second processor.

21. An apparatus for dynamic selection of a processor in a multiprocessor system, the apparatus comprising:

means for identifying a first interrupt from a communications adapter, the first interrupt including information identifying a separately addressable communications endpoint within the communications adapter;

means for identifying a routine for handling the first interrupt;

means for dispatching the first interrupt to a first processor upon checking a recent invocation indicator associated with the first processor to confirm that the first processor has a warm cache with respect to an interrupt handler corresponding to the first interrupt;

means for identifying a second interrupt;

means for dispatching the second interrupt to the first processor if a first interrupt characteristic and a second interrupt characteristic are different and if the first and second interrupts are associated with different scheduling groups; and

wherein the first and second characteristics identify a priority of the interrupt, an originating peripheral causing the interrupt, a type of a corresponding interrupt handler routine, the scheduling group, or whether the corresponding interrupt handler routine was recently run.

22. The apparatus of claim 21, wherein the second interrupt is dispatched to the first processor if the first and second interrupts correspond to messages of different protocols.

23. The apparatus of claim 21, wherein the second interrupt is dispatched to the first processor if the first and second interrupts are associated with different sessions.

24. The apparatus of claim 21, wherein the second interrupt is dispatched to the first processor if the first and second interrupts have different priorities.

25. The apparatus of claim 21, wherein the first processor is configured to run the routine for handling the interrupt in parallel with a second processor.

26. The apparatus of claim 21, wherein the apparatus is an interrupt controller.