Optimized buffer loading for packet header processing

• US 7,212,530 B1
• Filed: 03/22/2002
• Issued: 05/01/2007
• Est. Priority Date: 03/22/2002
• Status: Expired due to Fees

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: Juniper Networks Incorporated
• Inventors: Libby, Jeffrey G., Lim, Raymond M.
• Primary Examiner(s): Kizou; Hassan
• Assistant Examiner(s): SOL, ANTHONY M
• Application Number:
  US10/102,932
• Time in Patent Office: 1,866 Days
• Field of Search: None
• US Class Current: 370/392
• CPC Class Codes: H04L 49/3009 Header conversion, routing tables or routing tags
• CPC Class Codes: H04L 49/602 Multilayer or multiprotocol switching, e.g. IP switching

Abstract

A packet header processing engine includes a level 2 (L2) header generation unit and a level 3 (L3) header generation unit. The L2 and L3 header generation units are implemented in parallel with one another. The L2 generation unit writes L2 header information to a first buffer and the L3 generation unit writes L3 header information to a second buffer. When the L2 and L3 header generation units finish processing a packet, the packet may be unloaded from the first and second buffer while a new packet is simultaneously loaded to the packet header processing engine.

Claims

1. A packet header processing engine comprising:

an input memory configured to store incoming packet header information of a packet;

a packet processing unit configured to generate packet header information for the packet by operating on the packet header information stored in the input memory, the packet processing unit including a first processing component configured to generate layer 2 (L2) packet header information for the and a second processing component implemented in parallel with the first processing component and configured to generate layer 3 (L3) packet header information for the packet; and

a build component configured to receive the generated packet header information while the input memory stores incoming packet header information for a next packet.

2. The packet header processing engine of claim 1, further comprising:

a first output buffer configured to receive the packet header information generated by the first processing component; and

a second output buffer configured to receive the packet header information generated by the second processing component.

3. The packet header processing engine of claim 1, wherein the build component forms a combined packet header from the packet header information.

4. The packet header processing engine of claim 1, wherein the build component combines the generated packet header information relative to the first and second packet header protocols when the first and second processing components have completed processing the packet header information.

5. The packet header processing engine of claim 1, further comprising:

an instruction memory operatively coupled to the first processing component, the instruction memory providing the first processing component with programming instructions relating to the processing of the L2 packet header information.

6. The packet header processing engine of claim 1, further comprising:

an instruction memory operatively coupled to the processing component, the instruction memory providing the second processing component with programming instructions relating to the processing of the L3 packet header information.

7. The packet header processing engine of claim 6, wherein the second processing component includes:

a plurality of execution sections implemented in parallel with one another, each of the plurality of execution sections operating based on a portion of each instruction from the instruction memory.

8. The packet header processing engine of claim 1, further comprising:

registers connected to the first processing component and the second processing component.

9. A method comprising:

receiving packet header information for a packet;

generating, by a first packet processing unit, layer 2 (L2) protocol information for the packet based on the received packet header information and simultaneously generating, by a second packet processing unit, layer 3 (L3) protocol information for the packet based on the received packet header information;

receiving the L2 and L3 protocol information simultaneously while receiving packet header information for a next packet;

writing the generated L2 protocol information for the packet to a first buffer as the L2 protocol information is generated; and

writing the generated L3 protocol information for the packet to a second buffer as the L3 protocol information is generated.

10. The method of claim 9, wherein receiving the protocol information is performed when the generation of the protocol information for the packet is complete.

11. The method of claim 10, further comprising:

building a new packet header for the packet based on the protocol information.

12. The method of claim 11, further comprising:

combining the new packet header with payload information of the packet to form a complete packet.

13. A network device for processing packets comprising:

a buffer configured to store the packets, including header data and payload data for the packets;

a descriptor reader component coupled to the buffer, the descriptor reader component receiving the header data for the packets stored in the buffer and looking up descriptor information relating to the received packet header data; and

a packet header processing component coupled to the descriptor reader component, the packet header processing component receiving the packet header data and the descriptor information from the descriptor reader component, the packet header processing component additionally including

an input memory configured to store incoming packet header data and descriptor information of a packet,

a packet processing unit configured to generate, based on the packet header data and the descriptor information of the packet stored in the input memory, packet header data relating to a packet header protocol, the packet header processing unit including

a first processing component configured to generate layer 2 (L2) packet header data for the packet, and

a second processing component implemented in parallel with the first processing component and configured to generate layer 3 (L3) packet header data for the packet, and

a build component configured to receive the generated L2 and L3 packet header data while the input memory stores incoming packet header information for a next packet.

14. The network device of claim 13, further comprising:

a first output buffer configured to receive the L2 packet header data generated by the first packet processing component; and

a second output buffer configured to receive the L3 packet header data generated by the second packet processing component.

15. The network device of claim 13, wherein the first packet processing component is implemented in parallel with the second packet processing component.

16. The network device of claim 13, wherein the descriptor reader component looks up the descriptor information in external random access memory.

17. The network device of claim 13, wherein the network device is a network router.

18. The network device of claim 14, wherein the build component combines the generated packet header data from the first and second output buffers to form a combined packet header.

19. The network device of claim 18, wherein the build component combines the generated packet header information after the first packet processing unit and the second packet processing unit have completed generation of the L2 packet header data and the L3 packet header data, respectively.

20. The network device of claim 14, wherein the packet header processing component includes multiple groupings of the first packet processing component, the first output buffer, the second packet processing component, the second output buffer, and the build component, and wherein the multiple groupings implement parallel packet processing engines.

21. The network device of claim 13, wherein the packet header processing component additionally includes:

an instruction memory operatively coupled to the first packet processing component, the instruction memory providing the first packet processing component with programming instructions relating to the processing of the L2 packet header information.

22. The network device of claim 13, wherein the packet header processing component additionally includes:

an instruction memory operatively coupled to the second packet processing component, the instruction memory providing the second packet processing component with programming instructions relating to the processing of the L3 packet header information.

23. The network device of claim 22, wherein the second packet processing unit includes:

a plurality of execution sections implemented in parallel with one another, each of the plurality of execution sections operating based on a portion of each instruction from the instruction memory.

24. A device comprising:

means for receiving packet header information;

means for generating layer 2 (L2) protocol information for the packet based on the received packet header information;

means for writing the generated L2 protocol information for the packet as the L2 protocol information is generated;

means for generating layer 3 (L3) protocol information for the packet based on the received packet header information, the means for generating the L2 protocol information and the means for generating the L3 protocol information being implemented in parallel with one another;

means for writing the generated (L3) protocol information for the packet as the L3 protocol information is generated; and

means for unloading the written L2 and L3 protocol information while the means for receiving is receiving packet header information for a next packet.

25. The device of claim 24, wherein the means for writing the generated L2 protocol information writes the generated L2 protocol information to a first buffer and the means for writing the L3 protocol information writes the generated L3 protocol information to a second buffer.