METHOD AND APPARATUS FOR IMPROVING THE EFFICIENCY OF INTERRUPT DELIVERY AT RUNTIME IN A NETWORK SYSTEM

• US 20130173895A1
• Filed: 10/23/2012
• Published: 07/04/2013
• Est. Priority Date: 06/29/2007
• Status: Active Grant

• Current Assignee: Sujoy Sen, Yadong Li
• Original Assignee: Sujoy Sen, Yadong Li
• Inventors: Li, Yadong, Sen, Sujoy
• Granted Patent: US 8,838,864 B2
• Time in Patent Office: Days
• US Class Current: 712/244
• CPC Class Codes: G06F 13/24 using interrupt G06F13/32 takes precedence
• CPC Class Codes: G06F 9/3861 Recovery, e.g. branch miss-prediction, exception handling error detection or correction G06F11/00
• CPC Class Codes: H04L 49/90 Buffering arrangements

Abstract

Processor affinity of an application/thread may be used to deliver an interrupt caused by the application/thread to a best processor at runtime. The processor to which the interrupt is delivered may either run the target application/thread or be located in the same socket as the processor that runs the target application/thread. The processor affinity of the application/thread may be pushed down at runtime to a network device, a chipset, a memory control hub (“MCH”), or an input/output hub (“IOH”), which will facilitate delivery of the interrupt using that affinity information.

Claims

1. A method for improving the efficiency of interrupt delivery at runtime in a network system, comprising:

sending data to other users in the network system by an application;

forwarding the data to a device driver;

identifying a processor affinity of the application, the processor affinity including an identity of a processor that runs the application; and

associating an interrupt generated by a network device upon sending out the data with the processor affinity of the application.

2. The method of claim 1, further comprising delivering the interrupt directly to the processor that runs the application based at least in part on the processor affinity of the application.

3. The method of claim 1, wherein associating the interrupt with the processor affinity of the application comprises sending the processor affinity of the application to at least one of the network device or a chipset.

4. The method of claim 1, wherein associating the interrupt with the processor affinity of the application comprises selecting a transmit queue, the transmit queue associated with a message-based interrupt vector which includes the identity of the processor that runs the application.

5. The method of claim 4, further comprising:

sending the data through the selected transmit queue; and

transferring packets of the data from the selected transmit queue to the network device.

6. The method of claim 1, wherein the device driver comprises a driver of the network device, the network device including a network interface card.

7. The method of claim 6, further comprising generating an interrupt when the data is stored into a buffer of the network device or when the data is sent out of the network device.

8. The method of claim 1, wherein the device driver comprises an intermediate driver for a plurality of network devices.

9. An apparatus for improving the efficiency of interrupt delivery at runtime in a network system, comprising:

logic to

identify a processor affinity of an application, the processor affinity including an identity of a processor that runs the application; and

associate an interrupt generated by a network device upon sending out data from the application with the processor affinity of the application.

10. The apparatus of claim 9, wherein the logic further delivers the interrupt directly to the processor that runs the application based at least in part on the processor affinity of the application.

11. The apparatus of claim 9, wherein the logic associates the interrupt with the processor affinity of the application via operations including sending the processor affinity of the application to at least one of the network device or a chipset.

12. The apparatus of claim 9, wherein the logic associates the interrupt with the processor affinity of the application via operations including selecting a transmit queue, the transmit queue associated with a message-based interrupt vector which includes the identity of the processor that runs the application.

13. The apparatus of claim 12, wherein the logic further sends the data from the application through the selected transmit queue;

and transfers packets of the data from the selected transmit queue to the network device.

14. The apparatus of claim 9, wherein the logic comprises at least one of a driver of the network device or an intermediate driver for a plurality of network devices, the network device including a network interface card.

21. The computing system of claim 9, wherein the logic comprises at least one of a driver of the network device or an intermediate driver for a plurality of network devices.

15. A computing system, comprising:

logic to

identify a processor affinity of an application, the processor affinity including an identity of a processor that runs the application, and

associate an interrupt generated by a network device upon sending out data from the application with the processor affinity of the application; and

a circuit card coupled to a circuit board to host the network device, the circuit card operable to receive the data from the application and to send the data out to a destination.

16. The computing system of claim 15, wherein the logic further delivers the interrupt directly to the processor that runs the application based at least in part on the processor affinity of the application.

17. The computing system of claim 15, wherein the logic associates the interrupt with the processor affinity of the application via operations including sending the processor affinity of the application to at least one of the network device or a chipset.

18. The computing system of claim 15, wherein the logic associates the interrupt with the processor affinity of the application via operations including selecting a transmit queue, the transmit queue associated with a message-based interrupt vector which includes the identity of the processor that runs the application.

19. The computing system of claim 18, wherein the logic further sends the data from the application through the selected transmit queue;

and transfers packets of the data from the selected transmit queue to the network device.

20. The computing system of claim 18, wherein the circuit card operable to access a plurality of transmit queues including the selected transmit queue, and to transmit one or more packets from the plurality of transmit queues.

22. The computing system of claim 15, wherein the circuit card comprises a network interface card.

23. An article comprising a machine-readable medium that contains instructions, which when executed by a processing platform, cause said processing platform to perform operations including:

sending data to other users in the network system by an application;

forwarding the data to a device driver;

identifying a processor affinity of the application, the processor affinity including an identity of a processor that runs the application; and

associating an interrupt generated by a network device upon sending out the data with the processor affinity of the application.

24. The article of claim 23, wherein the operations further comprise delivering the interrupt directly to the processor that runs the application based at least in part on the processor affinity of the application.

25. The article of claim 23, wherein associating the interrupt with the processor affinity of the application comprises sending the processor affinity of the application to at least one of the network device or a chipset.

26. The article of claim 23, wherein associating the interrupt with the processor affinity of the application comprises selecting a transmit queue, the transmit queue associated with a message-based interrupt vector which includes the identity of the processor that runs the application.

27. The article of claim 26, wherein the operations further comprise:

sending the data through the selected transmit queue; and

transferring packets of the data from the selected transmit queue to the network device.

28. The article of claim 23, wherein the device driver comprises a driver of the network device, the network device including a network interface card.

29. The article of claim 23, wherein the operations further comprise generating an interrupt when the data is stored into a buffer of the network device or when the data is sent out of the network device.