User datagram protocol (UDP) receive offloading

• US 10,447,598 B2
• Filed: 07/17/2017
• Issued: 10/15/2019
• Est. Priority Date: 07/17/2017
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Snapy, Haim, Hedeshian, Harout, Desai, Pranav, Hershko, Yuval Corey
• Primary Examiner(s): Moore, Jr., Michael J
• Application Number:
  US15/651,944
• Publication Number:
  US 20190020587A1
• Time in Patent Office: 820 Days
• Field of Search: None
• CPC Class Codes: H04L 45/74 Address processing for routing
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 67/561 Adding application-functional data or data for application control, e.g. adding metadata
• CPC Class Codes: H04L 67/566 Grouping or aggregating service requests, e.g. for unified processing
• CPC Class Codes: H04L 69/02 Protocol performance
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers
• CPC Class Codes: H04W 28/0289 Congestion control load shedding arrangements in network planning H04W16/08; performing reselection for handling the traffic H04W36/22; wireless traffic scheduling H04W72/12

Abstract

Certain aspects of the present disclosure relate to methods and apparatus for providing UDP receive offloading using the communications systems. For example, a method generally includes grouping, into a container, the datasets from the same source and destination arriving at a receiver based on datasets metadata, and ungrouping the datasets from the container once network layer processing on the container is completed and datasets are ready to traverse to a next destination.

Claims

1. A method of processing datasets at a receiver, comprising:

receiving one or more packets;

grouping, into a container, datasets received in the one or more packets based on datasets metadata, comprising;

providing the container with a single header; and

ungrouping the datasets from the container once processing on the container is completed at a first processing layer of a network stack and datasets are ready to traverse to a second processing layer of the network stack.

2. The method of claim 1, wherein grouping the datasets comprises:

creating the container; and

dynamically determining the datasets that are to be grouped in the container based on dataset metadata.

3. The method of claim 2, wherein grouping datasets further comprises:

dynamically copy based or copy-less adding of the datasets into the container.

4. The method of claim 3, wherein grouping datasets further comprises:

setting the container to allow manipulation of the metadata on which a grouping mechanism relies.

5. The method of claim 1, wherein grouping the datasets comprises:

grouping, into the container, one or more of the packets based on the datasets metadata indicating that the multiple packets have similar headers.

6. The method of claim 5, wherein the single header is the same header as a first packet from the one or more packets added to the container.

7. The method of claim 1, wherein ungrouping datasets comprises:

determining the container to be ungrouped; and

dynamically removing datasets from the container while maintaining changes to the datasets made by network layer processing.

8. The method of claim 7, wherein ungrouping datasets further comprises:

discarding the container.

9. The method of claim 1, wherein ungrouping data sets comprises:

overwriting each original packet header within the container with the single header.

10. The method of claim 9, wherein ungrouping data sets further comprises:

retaining packet specific information including length from the original packet header.

11. The method of claim 10, wherein ungrouping data sets further comprises:

recalculating a checksum value to reflect any changes in the original packet header.

12. An apparatus for processing datasets by a receiver, comprising:

means for receiving one or more packets;

means for grouping, into a container, datasets received in the one or more packets based on datasets metadata, comprising;

means for providing the container with a single header; and

means for ungrouping the datasets from the container once processing on the container is completed at a first processing layer of a network stack and datasets are ready to traverse to a second processing layer of the network stack.

13. The apparatus of claim 12, wherein the means for grouping the datasets comprises:

means for creating the container; and

means for dynamically determining the datasets that are to be grouped in the container based on dataset metadata.

14. The apparatus of claim 13, wherein the means for grouping datasets further comprises:

means for dynamically copy based or copy-less adding of the datasets into the container.

15. The apparatus of claim 14, wherein the means for grouping datasets further comprises:

means for setting the container to allow manipulation of the metadata on which a grouping mechanism relies.

16. The apparatus of claim 12, wherein the means for grouping the datasets comprises:

means for grouping, into the container, one or more of the packets based on the datasets metadata indicating that the multiple packets have similar headers.

17. The apparatus of claim 16, wherein the single header is the same header as a first packet from the one or more packets added to the container.

18. The apparatus of claim 12, wherein the means for ungrouping datasets comprises:

means for determining the container to be ungrouped; and

means for dynamically removing datasets from the container while maintaining changes to the datasets made by network layer processing.

19. The apparatus of claim 18, wherein the means for ungrouping datasets further comprises:

means for discarding the container.

20. The apparatus of claim 12, wherein the means for ungrouping data sets comprises:

means for overwriting each original packet header within the container with the single header.

21. The apparatus of claim 20, wherein the means for ungrouping data sets further comprises:

means for retaining packet specific information including length from the original packet header.

22. The apparatus of claim 21, wherein the means for ungrouping data sets further comprises:

means for recalculating a checksum value to reflect any changes in the original packet header.

23. An apparatus for processing datasets at a receiver, comprising:

at least one processor configured to;

receive one or more packets;

group, into a container, datasets received in the one or more packets based on datasets metadata, comprising

to provide the container with a single header; and

ungroup the datasets from the container once processing on the container is completed at a first processing layer of a network stack and datasets are ready to traverse to a second processing layer of the network stack; and

a memory coupled to the at least one processor configured to store one or more of the received one or more packets, the container, and the datasets.

24. A non-transitory computer readable medium for processing datasets at a receiver having instructions stored thereon for:

receiving one or more packets;

grouping, into a container, datasets received in the one or more packets based on datasets metadata, comprising;

providing the container with a single header; and

ungrouping the datasets from the container once processing on the container is completed at a first processing layer of a network stack and datasets are ready to traverse to a second processing layer of the network stack.