Method and system for queuing a request by a processor to access a shared resource and granting access in accordance with an embedded lock ID

• US 8,918,791 B1
• Filed: 03/10/2011
• Issued: 12/23/2014
• Est. Priority Date: 03/10/2011
• Status: Active Grant

• Current Assignee: Ampere Computing LLC
• Original Assignee: Applied Micro Circuits Corporation (MACOM Technology Solutions Holdings, Inc.)
• Inventors: Sankaran, Kumar, Ravuri, Vinay, Chudgar, Keyur
• Primary Examiner(s): An, Meng
• Assistant Examiner(s): TEETS, BRADLEY A
• Application Number:
  US13/045,453
• Time in Patent Office: 1,384 Days
• Field of Search: None
• US Class Current: 718/104
• CPC Class Codes: G06F 9/526 Mutual exclusion algorithms

Abstract

A hardware-based method is provided for allocating shared resources in a system-on-chip (SoC). The SoC includes a plurality of processors and at least one shared resource, such as an input/output (IO) port or a memory. A queue manager (QM) includes a plurality of input first-in first-out memories (FIFOs) and a plurality of output FIFOs. A first application writes a first request to access the shared resource. A first application programming interface (API) loads the first request at a write pointer of a first input FIFO associated with the first processor. A resource allocator reads the first request from a read pointer of the first input FIFO, generates a first reply, and loads the first reply at a write pointer of a first output FIFO associated with the first processor. The first API supplies the first reply, from a read pointer of the first output FIFO, to the first application.

Claims

1. A method for queuing a request by a processor to access a shared resource, comprising:

receiving the request to access the shared resource, from a processor of a plurality of processors executing an application;

determining a queue order for a reply to the request based at least in part on the processor issuing the request and on a resource access priority level of the application executed by the processor;

generating the reply to the request comprising a grant to the shared resource with an embedded lock ID, wherein the embedded lock ID enables access to the shared resource for a predetermined length of time;

queuing the reply based on the queue order in an input of a first-in-first-out memory, wherein the input of the first in first out memory queues a set of replies corresponding to requests from the plurality of processors;

transmitting the reply to the processor;

receiving a second request to access the shared resource, from a second processor of the plurality of processors executing a second application;

denying access to the shared resource to the second processor requesting the shared resource in response to the predetermined length of time not having elapsed; and

generating a second reply to the second request that indicates that access was rejected and wherein the second reply also indicates an estimated time of availability for the shared resource.

2. The method of claim 1, wherein the reply comprises a grant with the embedded lock ID to a resource similar to the shared resource, a grant with the embedded lock ID and a use time limit, a reply with a resource availability estimate, a rejection, or combinations of the above-mentioned reply.

3. The method of claim 2, further comprising:

queuing the reply in a plurality of output first in first out memories, wherein each of the plurality of output first in first out memories queue replies for designated processors.

4. The method of claim 1 wherein receiving the request further comprises receiving a request comprising a go/no-go request for resource access, a scheduled time request for resource access, a length of time request for resource access, a resource access availability estimate, and a request for access to a resource similar to the shared resource, or combinations of the above-mentioned reply.

5. The method of claim 1 further comprising:

receiving a notification that the processor granted access to the shared resource has terminated the access to the shared resource; and

sending a notice to the second processor associated with the second request in a set of queued requests that the shared resource is available.

6. The method of claim 5 wherein sending the notice to the second processor associated with the second request comprises granting access to the shared resource.

7. The method of claim 1, further comprising granting access to a plurality of shared resources.

8. The method of claim 1, further comprising cycling allocation of access to the shared resource to the plurality of processors based on a predefined pattern.

9. The method of claim 1 further comprising cycling allocation of access to the shared resource based on an order in which the request was received.

10. A system that queues a reply to a request by a processor to access a shared resource, comprising:

a queue manager comprising an input of a first-in-first-out memory and an output of a first-in-first-out memory, wherein the queue manager is configured for receiving the request to access the shared resource, wherein the request is issued by a processor of a plurality of processors executing an application;

a resource allocator configured for granting access to the shared resource to a processor based on a queue order and is further configured for generating the reply to the request comprising a grant to the shared resource with an embedded lock ID, wherein the embedded lock ID enables access to the shared resource for a predetermined length of time;

the queue manager is also configured for queuing a reply to the request based on the queue order, wherein the queue order is determined based at least in part on the processor associated with the request and on a resource access priority level of the applications executed by the processor;

the queue manager is further configured for receiving a second request to access the shared resource, from a second processor of the plurality of processors executing a second application and the resource allocator is configured for denying the access to the shared resource to the second processor in response to the predetermined length of time not having elapsed; and

wherein the resource allocator is further configured for generating a second reply to the second request that indicates that access was rejected and wherein the second reply also indicates an estimated time of availability for the shared resource.

11. The system of claim 10 wherein the shared resource is selected from a group consisting of a memory and an input/out (IO) port.

12. The system of claim 10 wherein the resource allocator further comprises:

an allocation microprocessor; and

, an allocation application, executed by the allocation microprocessor, that is configured for tracking shared resource status and allocations.

13. The system of claim 10 wherein the reply generated by the resource allocator comprises a grant with the embedded lock ID and a use time limit a grant with the embedded lock ID to a resource similar to the shared resource, a reply with a resource availability estimate, a rejection, or combinations of the above mentioned reply.

14. The system of claim 10 wherein a set of requests are dequeued in a set of input first in first out memories and the replies are queued in a set of output first-in-first-out memories.

15. The system of claim 10 wherein the request comprises at least one request comprising a go/no-go request for resource access, a scheduled time request for resource access, a length of time request for resource access, a resource access availability estimate, and a request for access to a resource similar to the shared resource, or combinations of the above-mentioned requests.

16. The system of claim 10 wherein in response to receiving a notification that the processor granted access to the shared resource has terminated the access to the shared resource, the resource allocator is further configured for sending a notice to the second processor associated with a second request in a set of queued requests that the shared resource is available.

17. The system of claim 16 wherein the resource allocator is configured for sending the notification with a grant to access the shared resource.

18. The system of claim 10, wherein the resource allocator is configured for cycling allocation of access to the shared resource based on an order in which the request was received.

19. The system of claim 10, wherein the first in first out memory of the set of input and output first in first out memories are associated with designated processors.

20. A non-transitory computer-readable storage device comprising computer-executable instructions that, in response to execution, cause a system including a processor to perform operations, comprising:

receiving a request to access a shared resource, wherein the request is issued by a processor of a plurality of processors executing an application;

determining a queue order for a reply to the request based at least in part on the processor and on a resource access priority level of the application executed by the processor;

generating the reply to the request comprising a grant to the shared resource with an embedded lock ID, wherein the embedded lock ID enables access to the shared resource for a predetermined length of time;

queuing the reply request in an input first-in-first-out memory based on the queue order, wherein the input first-in-first-out memory queues replies to a plurality of designated processors;

transmitting the reply to the processor;

receiving a second request to access the shared resource, from a second processor of the plurality of processors executing a second application;

denying access to the shared resource to the second processor requesting the shared resource in response to the predetermined length of time not having elapsed; and

generating a second reply to the second request that indicates that access was rejected and wherein the second reply also indicates an estimated time of availability for the shared resource.