Data routing acceleration

• US 8,848,713 B2
• Filed: 10/13/2009
• Issued: 09/30/2014
• Est. Priority Date: 10/13/2009
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: DiMambro, Francesco
• Primary Examiner(s): Blanton, John
• Assistant Examiner(s): PHAM, TITO Q
• Application Number:
  US12/578,181
• Publication Number:
  US 20110085549A1
• Time in Patent Office: 1,813 Days
• Field of Search: None
• US Class Current: 370/392
• CPC Class Codes: H04L 12/462 LAN interconnection over a bridge based backbone
• CPC Class Codes: H04L 45/745 Address table lookup; Address filtering
• CPC Class Codes: H04L 49/9094 Arrangements for simultaneous transmit and receive, e.g. simultaneous reading/writing from/to the storage element
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/161 Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level

Abstract

Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. Additionally, the method includes passing the first packet to one of a bridge and IP layer of the receive stack. A first packet inhibitor at the one of the bridge and IP layer is bypassed upon recognizing the first packet as a first packet type. The first packet is routed to a transmit stack of the electronic network device from the one of the bridge and IP layer.

Claims

1. A method of processing data packets comprising:

receiving data packets at a receive stack at an ingress interface of an electronic network device, wherein the receive stack includes a first application layer, a first transport control protocol (TCP) layer, a first Internet protocol (IP) layer, and a first data link layer, the first data link layer having a first bridge layer and a first firmware layer herein;

creating within the first firmware layer a first packet from at least a portion of each of the set of data packets, the first packet including an indicator that the first packet is a large receive offload (LRO) packet;

passing the first packet from the first firmware layer to the first bridge of the receive stack;

upon recognizing the first packet as an LRO packet based on the indicator, overriding a packet size limitation implemented by the first bridge, thereby enabling the first bridge to process the first packet;

determining whether the first packet matches a media access control (MAC) address in a MAC address table provided by the first bridge;

if no match is found in the MAC address table, determining whether the first packet matches an IP address in an IP address table provided by the first IP layer;

if a match is found, transmitting the first packet to a corresponding layer of a transmit stack of an egress interface of the electronic network device without passing the first packet to the first TCP layer or the first application layer of the receive stack, wherein the transmit stack includes a second application layer, a second transport control protocol (TCP) layer, a second Internet protocol (IP) layer, and a second data link layer, the second data link layer having a second bridge layer and a second firmware layer herein, wherein the corresponding layer of the transmit stack that receives the first packet is one of the second bridge layer and the second firmware layer; and

routing, by the transmit stack of the egress interface, the first packet to a destination specified by the match via the egress interface.

2. The method of claim 1 further comprising:

receiving the first packet at the transmit stack; and

providing an indicator that the first packet is a second packet type.

3. The method of claim 1 wherein providing an indicator that first packet is an LRO packet comprises setting one or more bits in a header of the first packet.

4. The method of claim 1 wherein creating the first packet from at least a portion of each of the set of data packets comprises:

creating a first packet header from one of the received packets in a buffer; and

aggregating non-header data from all of the received packets in the buffer with the header.

5. The method of claim 4 further comprising providing the first packet to a driver in the transmit stack, the driver being configured to:

break the first packet into packet segments;

create headers for the packet segments using the first packet header as a template;

generate checksums for the packet segments; and

encapsulate the packet segments, wherein encapsulation comprises;

appending packet segments with created headers; and

providing a frame header and footer.

6. The method of claim 1 further comprising:

determining a size of each of the received packets; and

setting a packet size limit to correspond with the size of the largest received packet, if the largest received packet exceeds a current packet size limit.

7. The method of claim 1 further comprising:

receiving the first packet at the transmit stack;

segmenting the first packet into smaller packets; and

providing the smaller packets to a driver in the transmit stack.

8. The method of claim 1 further comprising:

generating a packet map correlative with the large packet;

providing the packet map to the transmit stack; and

using the packet map to transmit the data of the first packet.

9. An electronic network device comprising:

a processor;

a memory coupled to the processor, the memory storing operating instructions for the operation of the network device;

a transmit stack and a receive stack as part of a network stack provided by an operating system operated by the processor, the transmit stack being associated with an egress interface of the electronic network device and the receive stack being associated with an ingress interface of the electronic network device, each of the transmit stack and receive stack comprising;

a first layer being a data link layer comprising;

a driver; and

a bridge layer logically adjacent to the driver and operated by the processor, the bridge layer including a MAC address table for mapping media control access (MAC) addresses;

a second layer being an Internet protocol (IP) layer logically adjacent to the bridge layer, the second layer including an IP address table for mapping IP addresses;

a transport control protocol (TCP) layer coupled to the IP layer; and

an application layer coupled to the TCP layer,

wherein a driver of the receive stack is configured to create a first packet of a large receive offload (LRO) packet type from a plurality of received data packets, the first packet including an indication for overriding packet size limits;

wherein the bridge layer of the receive stack is configured to;

recognize the indication for overriding packet size limits;

overriding a packet size limitation implemented by the bridge layer of the receive stack, thereby enabling the bridge layer of the receive stack to process the first packet;

determining whether the first packet matches a media access control (MAC) address in the MAC address table of the receive stack; and

if a match is found, forward the first packet to a bridge layer of the transmit stack upon finding a common address in the MAC address table of the receive stack and a MAC address table of the transmit stack, without forwarding the first packet to the TCP layer or the application layer of the receive stack;

wherein the IP layer of the receive stack is configured to;

if no match is found in the MAC address table of the receive stack, determine whether the first packet matches an IP address in an IP address table of the receive stack; and

if a match is found, forward the first packet to an IP layer of the transmit stack upon finding a common address in the IP address table of the receive stack and an IP address table of the transmit stack, without forwarding the first packet to the TCP layer or the application layer of the receive stack.

10. The electronic network device of claim 9 wherein the second layer of the receive stack is configured to recognize the indication for overriding packet size limits and forward the first packet to a second layer of the transmit stack upon finding a common address in the respective second address tables of the receive stack and the transmit stack.

11. The electronic network device of claim 10 wherein the bridge layer and the second layer of the receive stack are configured to indicate the first packet as being a second type upon forwarding the first packet to the transmit stack.

12. The electronic network device of claim 11 wherein the bridge and second layers of the transmit stack are configured to forward the first packet to the driver upon recognizing the first packet as a second type, driver of the transmit stack being configured to segment the first packet in preparation of transmission.

13. The electronic network device of claim 11 wherein the bridge layer and the second layer of the transmit stack are configured to segment the first packet upon recognizing the first packet as a second type.

14. The electronic network device of claim 9 wherein the driver of the transmit stack is configured to communicate segmentation capabilities to the bridge and second layers of the transmit stack.

15. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform a method for operating an electronic network device, the method comprising:

receiving data packets at a receive stack at an ingress interface of the electronic network device, wherein the receive stack includes a first application layer, a first transport control protocol (TCP) layer, a first Internet protocol (IP) layer, and a first data link layer, the first data link layer having a first bridge layer and a first firmware layer herein;

creating within the first firmware layer a first packet from at least a portion of each of the set of data packets, the first packet including an indicator that the first packet is a large receive offload (LRO) packet;

passing the first packet from the first firmware layer to the first bridge of the receive stack;

upon recognizing the first packet as an LRO packet based on the indicator, overriding a packet size limitation implemented by the first bridge, thereby enabling the first bridge to process the first packet;

determining whether the first packet matches a media access control (MAC) address in a MAC address table provided by the first bridge;

if no match is found in the MAC address table, determining whether the first packet matches an IP address in an IP address table provided by the first IP layer;

if a match is found, transmitting the first packet to a corresponding layer of a transmit stack of an egress interface of the electronic network device without passing the first packet to the first TCP layer or the first application layer of the receive stack, wherein the transmit stack includes a second application layer, a second transport control protocol (TCP) layer, a second Internet protocol (IP) layer, and a second data link layer, the second data link layer having a second bridge layer and a second firmware layer herein, wherein the corresponding layer of the transmit stack that receives the first packet is one of the second bridge layer and the second firmware layer; and

routing, by the transmit stack of the egress interface, the first packet to a destination specified by the match via the egress interface.

16. The non-transitory machine-readable medium of claim 15 wherein the method further comprises providing the first packet to the transmit stack via a bridge layer or a IP layer.

17. The non-transitory machine-readable medium of claim 16 wherein the method further comprises converting the first packet to a second packet type.

18. The non-transitory machine-readable medium of claim 16 wherein the method further comprises segmenting the large receive offload packet into smaller packets at driver of the transmit stack.