Method and apparatus for pre-classifying packets

• US 10,462,267 B2
• Filed: 05/31/2016
• Issued: 10/29/2019
• Est. Priority Date: 03/21/2011
• Status: Active Grant

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: Marvell World Trade Limited (Marvell Technology Group Limited)
• Inventors: Zeffer, Hakan, Carlstrom, Jakob, Westlund, Par, Back, Johan, Nilsson, Ronny
• Primary Examiner(s): Beharry, Noel R
• Assistant Examiner(s): Preval, Lionel
• Application Number:
  US15/169,011
• Publication Number:
  US 20160277549A1
• Time in Patent Office: 1,246 Days
• Field of Search: 370392, 370408, 398 49
• CPC Class Codes: H04L 45/70 Routing based on monitoring results
• CPC Class Codes: H04L 45/74 Address processing for routing
• CPC Class Codes: H04L 47/2441 relying on flow classification, e.g. using integrated services [IntServ]
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers

Abstract

The disclosed embodiments relate to a system that provides an intelligent port infrastructure for communication network devices. This is accomplished by incorporating a highly configurable pre-classifier module into the port infrastructure. This pre-classifier makes it possible to realign packet data to add a configurable number of bytes to the front of the packet, and also to select interesting data from incoming packets for further analysis. The selected data is sent into a configurable classification engine, which generates instructions that specify how to determine associated packet attributes. The packet attributes are then generated based on the instructions, and are forwarded along with the packet to downstream processing units.

Claims

1. A method for processing a packet, the method comprising:

receiving the packet at a classifier from a plurality of input ports connected to the classifier;

performing a classification operation based upon data selected from within the packet and data other than the data selected from within the packet;

generating attributes for the packet based upon a result of the classification operation; and

forwarding the attributes along with the packet to facilitate subsequent packet-processing operations,

wherein the data selected from within the packet is selected from a set of locations within the packet, the data selected being associated with the input port from which the packet was received from, and

wherein the attributes include a destination, and wherein (i) a default value for the destination is a per port defined value, and (ii) the destination is a sum of (a) the result of the classification operation, (b) the default value, and (c) a channel number associated with the packet.

2. The method of claim 1, wherein receiving the packet includes adding bytes to a front of the packet.

3. The method of claim 2, wherein a configurable number of the bytes to add to the front of the packet is determined based upon the input port the packet was received from.

4. The method of claim 1, wherein a number of the set of locations is nine or fewer.

5. The method of claim 1, wherein the classification operation includes performing a lookup in a lookup structure based on the data selected from within the packet.

6. The method of claim 5, wherein the lookup structure is a ternary content-addressable memory (TCAM).

7. The method of claim 6, wherein generating attributes for the packet includes combining data from the result of the classification operation with at least one of i) data from the packet and ii) information forwarded from the input port the packet was received from.

8. The method of claim 1, wherein the data other than the data selected from within the packet includes at least one of i) the channel number and ii) a programmable identifier associated with the input port the packet was received from.

9. The method of claim 8, wherein the at least one of the channel number and the programmable identifier are configurable on a per-port basis.

10. The method of claim 9, wherein the at least one of the channel number and the programmable identifier are received from the input port the packet was received from.

11. The method of claim 8, wherein the programmable identifier identifies a particular group of ports in the classification operation.

12. A packet processing apparatus, the apparatus comprising:

a circuit configured to;

receive a packet at a classifier from a plurality of input ports connected to the classifier;

perform a classification operation based upon data selected from within the packet and data other than the data selected from within the packet;

generate attributes for the packet based upon a result of the classification operation; and

forward the attributes along with the packet to facilitate subsequent packet-processing operations, wherein

the data selected from within the packet is selected from a set of locations within the packet, the data selected being associated with the input port from which the packet was received from, and

wherein the attributes include a destination, and wherein (i) a default value for the destination is a per port defined value, and (ii) the destination is a sum of (a) the result of the classification operation, (b) the default value, and (c) a channel number associated with the packet.

13. The apparatus of claim 12, wherein the circuit is further configured to add bytes to a front of the packet.

14. The apparatus of claim 13, wherein the circuit is further configured to determine a configurable number of the bytes to add to the front of the packet based upon the input port the packet was received from.

15. The apparatus of claim 12, wherein a number of the set of locations is nine or fewer.

16. The apparatus of claim 15, wherein the circuit is further configured to perform a lookup in a lookup structure based on the data selected from within the packet.

17. The apparatus of claim 16, wherein the lookup structure is a ternary content-addressable memory (TCAM).

18. The apparatus of claim 17, wherein the circuit is further configured to combine data from the result of the classification operation with at least one of i) data from the packet and ii) information forwarded from the input port the packet was received from.

19. The apparatus of claim 12, wherein the data other than the data selected from within the packet includes at least one of i) the channel number and ii) a programmable identifier associated with the input port the packet was received from.

20. The apparatus of claim 19, wherein the at least one of the channel number and the programmable identifier are configurable on a per-port basis.

21. The apparatus of claim 20, wherein the at least one of the channel number and the programmable identifier are received from the input port the packet was received from.

22. The apparatus of claim 19, wherein the programmable identifier identifies a particular group of ports in the classification operation.

23. A system for processing a packet, the system comprising:

a fabric switch; and

a plurality of lines cards communicably coupled to the fabric switch, wherein

at least one of the line cards includes a network processor configured to;

receive the packet at a classifier from a plurality of input ports connected to the classifier;

perform a classification operation based upon data selected from within the packet and data other than the data selected from within the packet;

generate attributes for the packet based upon a result of the classification operation; and

forward the attributes along with the packet to facilitate subsequent packet-processing operations,

the data selected from within the packet is selected from a set of locations within the packet, the data selected being associated with the input port from which the packet was received from, and

wherein the attributes include a destination, and wherein (i) a default value for the destination is a per port defined value, and (ii) the destination is a sum of (a) the result of the classification operation, (b) the default value, and (c) a channel number associated with the packet.