Dynamic allocation of buffers for direct memory access

• US 8,732,352 B1
• Filed: 01/25/2013
• Issued: 05/20/2014
• Est. Priority Date: 01/25/2013
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Herbeck, Gilbert
• Primary Examiner(s): TSENG, CHENG YUAN
• Application Number:
  US13/750,266
• Time in Patent Office: 480 Days
• Field of Search: None
• US Class Current: 710/22
• CPC Class Codes: G06F 13/385 for adaptation of a particular data processing system to different peripheral devices
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A device includes a processor, unified DMA (‘Direct Memory Access’) storage, and a number of DMA engines. The processor may be operatively coupled to the unified DMA storage and a main memory. The DMA engines may be configured to access the unified DMA storage and provide DMA transmissions between the main memory and a corresponding component. The processor may be configured to: determine a size of a corresponding DMA buffer to be allocated for each DMA engine; allocate, for each DMA engine, the corresponding DMA buffer of the determined size in the unified DMA storage; and execute DMA transmission using the DMA engines and the corresponding DMA buffers.

Claims

1. A device comprising:

unified DMA (‘

Direct Memory Access’

) storage;

a processor operatively coupled to the unified DMA storage and a main memory; and

a plurality of DMA (‘

Direct Memory Access’

) engines, wherein the plurality of DMA engines are configured to access the unified DMA storage and provide DMA transmissions between the main memory and a corresponding component; and

wherein the processor is configured to;

determine a size of a corresponding DMA buffer to be allocated for each DMA engine;

allocate, for each DMA engine, the corresponding DMA buffer of the determined size in the unified DMA memory; and

execute DMA transmissions using the DMA engines and the corresponding DMA buffers.

2. The device of claim 1, wherein the processor is further configured to:

monitor DMA traffic between the DMA engines and the corresponding components; and

reallocate one or more corresponding DMA buffers for the DMA engines in dependence upon the monitored traffic.

3. The device of claim 1, wherein the processor is further configured to determine that no corresponding DMA buffer be allocated for at least one DMA engine.

4. The device of claim 1, wherein the processor is further configured to determine the size of the corresponding DMA buffer to be allocated for each DMA engine in dependence upon expected DMA bandwidth utilization between the DMA engines and the corresponding components.

5. The device of claim 1 wherein the device further comprises a wireless mobile device.

6. The device of claim 1, wherein the unified DMA storage comprises an array of physical memory cells.

7. The device of claim 1, wherein the processor is further configured to:

determine a first size for a corresponding transmit DMA buffer for each DMA engine;

determine a second size for a corresponding receive DMA buffer for each DMA engine;

allocate, for each DMA engine, the corresponding transmit DMA buffer of the first determined size in the unified DMA storage; and

allocate, for each DMA engine, the corresponding receive DMA buffer of the second determined size in the unified DMA storage.

8. A method of dynamically administering DMA (‘

Direct Memory Access’

) buffers for a plurality of DMA engines, wherein each DMA engine is configured to provide DMA transmissions between a main memory and a corresponding component, the method comprising;

determining, by the processor, a size of a corresponding DMA buffer to be allocated for each DMA engine;

allocating, by the processor for each DMA engine, the corresponding DMA buffer of the determined size in unified DMA storage accessible by the DMA engines; and

executing DMA transmissions using the DMA engines and corresponding DMA buffers.

9. The method of claim 8, further comprising:

monitoring DMA traffic between the DMA engines and corresponding components; and

reallocating one or more corresponding DMA buffers for the DMA engines in dependence upon the monitored traffic.

10. The method of claim 8, wherein determining, by a processor, a size of a corresponding DMA buffer to be allocated for each DMA engine further comprises determining that no corresponding DMA buffer be allocated for at least one DMA engine.

11. The method of claim 8, wherein determining, by a processor, a size of a corresponding DMA buffer to be allocated for each DMA engine further comprises determining the size in dependence upon expected DMA bandwidth utilization between the DMA engines and the corresponding components.

12. The method of claim 8, wherein the unified DMA storage comprises an array of physical memory cells.

13. A processor for dynamically administering DMA (‘

Direct Memory Access’

) buffers for a plurality of DMA engines, wherein each DMA engine is configured to provide DMA transmission between a main memory and a corresponding component, wherein the processor is configured to;

determine a size of a corresponding DMA buffer to be allocated for each DMA engine;

allocate, for each DMA engine, the corresponding DMA buffer of the determined size in unified DMA storage accessible by the DMA engines; and

execute DMA transmissions using the DMA engines and the corresponding DMA buffers.

14. The processor of claim 13, further configured to:

monitor DMA traffic between the DMA engines and corresponding components; and

reallocate one or more corresponding DMA buffers for the DMA engines in dependence upon the monitored traffic.

15. The processor of claim 13, wherein the processor is further configured to determine that no corresponding DMA buffer be allocated for at least one DMA engine.

16. The processor of claim 13, wherein the processor is further configured to determine the size of the corresponding DMA buffer to be allocated for each DMA engine in dependence upon expected DMA bandwidth utilization between the DMA engines and the corresponding components.

17. The processor of claim 13, wherein the processor, the main memory, the unified DMA storage, and the DMA engines comprise components of a system-on-chip.

18. The processor of claim 17, wherein one or more of the DMA engines is configured to provide DMA transmission between the main memory and a corresponding on-chip component.

19. The processor of claim 17, wherein one or more of the DMA engines is configured to provide DMA transmissions between the main memory and a corresponding off-chip component.

20. The processor of claim 13, wherein the unified DMA storage comprises an array of physical memory cells.

21. A computer readable storage medium including computer program instructions for dynamically administering DMA (‘

Direct Memory Access’

) buffers for a plurality of DMA engines, wherein each DMA engine is configured to provide DMA transmissions between a main memory and a corresponding component and the computer program instructions are executable by the processor to;

determine a size of a corresponding DMA buffer to be allocated for each DMA engine;

allocate, for each DMA engine, the corresponding DMA buffer of the determined size in unified DMA storage accessible by the DMA engines; and

execute DMA transmissions using the DMA engines and the corresponding DMA buffers.

22. The computer readable storage medium of claim 21, further comprising computer program instructions executable by the processor to:

monitor DMA traffic between the DMA engines and the corresponding components; and

reallocate one or more corresponding DMA buffers for the DMA engines in dependence upon the monitored traffic.

23. The computer readable storage medium of claim 21, wherein the computer program instructions executable by the processor to determine a size of a corresponding DMA buffer to be allocated for each DMA engine further comprise computer program instructions executable by the processor to determine that no corresponding DMA buffer be allocated for at least one DMA engine.

24. The computer readable storage medium of claim 21, wherein the computer program instructions executable by the processor to determine a size of a corresponding DMA buffer to be allocated for each DMA engine further comprise computer program instructions executable by the processor to determine the size in dependence upon expected DMA bandwidth utilization between the DMA engines and the corresponding components.

25. The computer readable storage medium of claim 21, wherein the unified DMA storage comprises an array of physical memory cells.