System for combining requests associated with one or more memory locations that are collectively associated with a single cache line to furnish a single memory operation

• US 6,434,639 B1
• Filed: 11/13/1998
• Issued: 08/13/2002
• Est. Priority Date: 11/13/1998
• Status: Expired due to Term

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Haghighi, Siamack
• Primary Examiner(s): Gaffin, Jeffrey
• Assistant Examiner(s): Perveen, Rehana
• Application Number:
  US09/191,923
• Time in Patent Office: 1,369 Days
• Field of Search: 710/35, 710/39, 711/118, 711/167, 711/100, 711/3
• US Class Current: 710/39
• CPC Class Codes: G06F 12/0831 using a bus scheme, e.g. with bus monitoring or watching means

Abstract

A method and apparatus are for use with a computer system. Write requests to store data in one or more memory locations that are collectively associated with a cache line are received. The first requests are combined to furnish a memory operation. The computer system may include a peripheral device that furnishes a stream of data to be stored in a memory, and the apparatus may include first and second interfaces, a queue and logic. The first interface is adapted to convert a portion of the stream of data into a first request, and the queue is adapted to store the second request. The logic is adapted to determine if the first and second requests target memory locations that are collectively associated with a cache line and based on the determination, selectively combine the first and second requests. The second interface is adapted to furnish a memory operation in response to the combination of the first and second requests. The requests may be, as examples, read requests and/or write requests.

Claims

1. A computer system comprising:

a memory including memory locations that are collectively associated with a cache line;

a peripheral device adapted to furnish a stream of data; and

bridge circuitry coupled between the memory and the peripheral device, the bridge circuitry comprising;

a first interface adapted to convert a portion of the stream of data into a first write request;

a queue adapted to store a second request;

logic adapted to determine if the first and second requests target memory locations are collectively associated with a cache line and based on the determination, selectively combine the first and second requests; and

a second interface adapted to furnish a memory operation in response to the combination of the first and second requests by the logic.

2. The computer system of claim 1, wherein the bridge circuitry comprises a at least a portion of a bridge.

3. The computer system of claim 1, wherein the bridge circuitry comprises at least portions of two bridges.

4. The computer system of claim 1, wherein the peripheral device comprises a camera.

5. The computer system of claim 1, wherein the peripheral device comprises a block oriented device.

6. The computer system of claim 5, wherein the peripheral device comprises a disk drive.

7. The computer system of claim 1, wherein the queue comprises:

a first memory adapted to store the second request; and

a second memory adapted to store data associated with the second request.

8. The computer system of claim 1, wherein the first and second requests comprise write requests.

9. The computer system of claim 1, wherein the first and second requests comprise read requests.

10. An apparatus for use with a computer system including a peripheral device that furnishes a stream of data to be stored in a memory, the apparatus comprising:

a first interface adapted to convert a portion of the stream of data into a first write request;

a queue adapted to store a second write request;

logic adapted to determine if the first and second write requests target memory locations that are collectively associated with a cache line and based on the determination, selectively combine the first and second write requests; and

a second interface adapted to furnish a write operation to the memory locations in response to the combination of the first and second write requests.

11. The apparatus of claim 10, wherein the apparatus comprises at least a portion of a bridge chip.

12. The apparatus of claim 10, wherein the queue comprises:

a first memory adapted to store the second request; and

a second memory adapted to store data associated with the second request.

13. A method for use with a computer system, comprising:

receiving first requests associated with one or more locations of a first memory, the memory locations being collectively associated with a cache line of a cache memory separate from the first memory; and

combining the first requests to furnish an operation associated with the first memory.

14. The method of claim 13, wherein the first requests comprise write requests and the operation comprises a write operation.

15. The method of claim 13, wherein the first requests comprise read requests and the operation comprises a read operation.

16. The method of claim 13, further comprising:

receiving a data stream; and

generating the first requests from the data stream.

17. The method of claim 13, wherein the first requests comprise write requests, further comprising;

prefetching data associated with at least the cache line based on at least one of the write requests.

18. The method of claim 13, wherein the act of combining comprises:

linking the first requests together.

19. The method of claim 13, wherein the act of combining comprises:

generating a request from the first requests.

20. A method comprising:

receiving a first request;

determining if the first request and a second request stored in a queue target location of a first memory, the requests being collectively associated with a cache line of a cache memory separate loom the first memory; and

based on the determination combining the first and second requests to generate a an operation associated with the first memory.

21. The method of claim 20, wherein the first requests comprise write requests and the operation comprises a write operation.

22. The method of claim 20, wherein the first requests comprise read requests and the operation comprises a read operation.

23. The method of claim 20, wherein the act of combining comprises:

linking the first and second requests together.

24. The method of claim 20, wherein the act of combining comprises:

generating a third request from the first and second requests.

25. The method of claim 24, further comprising:

storing the third write request in the queue.

26. The method of claim 20, further comprising:

retrieving a stream of data from a peripheral device; and

generating the first request from a portion of the stream of data.

27. The method of claim 26, wherein the peripheral device comprises a camera.

28. The method of claim 26, wherein the peripheral device comprises a block oriented device.

29. The method of claim 26, wherein the peripheral device comprises a disk drive.