SYSTEMS AND METHOD FOR TRANSACTION STALL DETECTION AND PROPAGATING THE RESULT IN A MULTI-CORE ARCHITECTURE

• US 20100325495A1
• Filed: 06/18/2010
• Published: 12/23/2010
• Est. Priority Date: 06/22/2009
• Status: Active Grant

• Current Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Original Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Inventors: Chauhan, Abhishek, Talla, Ramanjaneyulu Y., Bots, Henk
• Granted Patent: US 8,381,026 B2
• Time in Patent Office: Days
• US Class Current: 714/49
• CPC Class Codes: G06F 11/0712 in a virtual computing platform, e.g. logically partitioned systems
• CPC Class Codes: G06F 11/0724 in a multiprocessor or a multi-core unit multiprocessors per se G06F15/80
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer

Abstract

The present invention is directed towards systems and methods for determining failure in and controlling access to a shared resource in a multi-core system. In some embodiments of a multi-core system, individual cores may share the same resource. Additionally, the resource may occasionally fail or need to be reset, and the period during which the resource is being reset may be non-instantaneous. In an embodiment without coordination between the cores, one core experiencing a failure may reset the resource. During the period in which the resource is resetting, another core may interpret the reset as a failure and reset the resource. As more cores interpret the resets as failures, they will trigger resets, quickly resulting in the resource being constantly reset and unavailable. Thus, in some embodiments, a coordination system may be utilized to determine failure of a shared resource and control resets and access to the shared resource.

Claims

1. A method for managing an error with a resource shared among cores in a multi-core device, the method comprising:

(a) receiving, by an interface master executing on a multi-core device, a first message indicating an error with a resource shared by cores of the multi-core device;

(b) communicating, by the interface master, to each of one or more interface slaves, each executing on a corresponding core of the multi-core device, a second message indicating the error with the shared resource, the corresponding core halting use of the shared resource responsive to the second message;

(c) resetting, by the interface master responsive to the first message, the shared resource; and

(d) communicating, by the interface master responsive to determining the resetting of the shared resource is complete, to each interface slave, a third message indicating the shared resource is available.

2. The method of claim 1, wherein step (a) further comprises receiving, by the interface master, the first message indicating a transmission stall of one or more transmission queues for network communications.

3. The method of claim 1, wherein step (a) further comprises receiving, by the interface master, the first message from an interface slave.

4. The method of claim 1, wherein step (a) further comprises tracking, by the interface master, a number of times the error has occurred within a predetermined time period.

5. The method of claim 1, wherein step (b) further comprising communicating, by the interface master, the second message as an inter-core message.

6. The method of claim 1, wherein step (b) further comprises resetting, by each interface slave responsive to the second message, one or more resources local to the corresponding core and associated with the shared resource.

7. The method of claim 1, wherein step (b) further comprises determining, by the interface master, that the error has occurred a predetermined number of times within a predetermined timer period, and responsive to the determination, communicating to each interface slave, the second message indicating the shared resource is disabled.

8. The method of claim 1, wherein step (b) further comprises determining, by the interface master, that the error has not occurred a predetermined number of times within a predetermined timer period, and responsive to the determination, communicating to each interface slave, the second message indicating the shared resource is being reset.

9. The method of claim 1, wherein step (d) further comprises determining, by the interface matter that the shared resource has been successfully reset.

10. The method of claim 1, wherein step (c) further comprises communicating, by the interface master, a request to reset the shared resource comprising a network interface card.

11. A method for handling a transmission stall for a network interface resource shared by a plurality of cores in a multi-core device, the method comprising:

(a) identifying, by an interface master executing on a multi-core device, that a transmission stall has occurred with a network interface resource shared by cores of the multi-core device;

(b) determining, by the interface master, a predetermined number of transmission stalls have occurred with a predetermined time period; and

(c) communicating, by the interface master responsive to the determination, to each of one or more interface slaves, each executing on a corresponding core of the multi-core device, a message indicating the network interface resource is disabled.

12. The method of claim 11, wherein step (a) further comprises detecting, by an interface slave on a core, the transmission stall of the network interface resource, and sending a message to the interface master identifying the transmission stall.

13. The method of claim 11, wherein step (a) further comprises detecting, by the interface master, the transmission stall of the network interface resource.

14. The method of claim 11, wherein step (a) further comprises detecting, by a core of the multi-core device, the transmission stall via monitoring of one or more transmission buffers of the network interface resource.

15. The method of claim 11, wherein step (b) further comprises identifying, via the multi-core device, the predetermined number of transmission stalls via a policy.

16. The method of claim 11, wherein step (b) further comprises identifying, via the multi-core device, the predetermined time period via a policy.

17. The method of claim 11, wherein step (c) further comprises communicating, by the interface master, the message via a unicast inter-core message.

18. The method of claim 11, further comprising resetting, by the interface master responsive to the first message, the network interface resource.

19. The method of claim 11, further comprising re-initializing, by the interface master responsive to the first message, the network interface resource.

20. The method of claim 11, communicating, by the interface master, to each interface slave, a second message indicating the shared resource is enabled.