Compact packet operation device and method

• US 20070115966A1
• Filed: 06/01/2006
• Published: 05/24/2007
• Est. Priority Date: 11/21/2005
• Status: Active Grant

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.)
• Inventors: Tzeng, Shrjie
• Granted Patent: US 7,869,411 B2
• Time in Patent Office: Days
• US Class Current: 370/389
• CPC Class Codes: H04L 45/742 Route cache; Operation thereof
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/2458 Modification of priorities while in transit
• CPC Class Codes: H04L 47/32 by discarding or delaying data units, e.g. packets or frames
• CPC Class Codes: H04L 63/0227 Filtering policies mail message filtering H04L51/212
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers

Abstract

A network device and a method of handling a frame in a network device defining a common header for each rule in a rules table of a search engine, wherein the common header comprises a source port specifying a corresponding rule to be applied to a port or ports, and a frame format specifying a format applied to the corresponding rule. A port receives a frame, with the frame having multiple field values. A pre-parser parses the received frame to obtain the field values and corresponding slices. A search engine receives the slices and determines matches between the slices and the common header of each rule in the rules table. A policy table performs indexing based on the determined matches to obtain an action entry. An action resolution takes an action based on the obtained action entry.

Claims

1. A method of handling a frame in a network device, the method comprising:

defining a common header for each rule in a rules table of a search engine, wherein the common header comprises a source port specifying a corresponding rule to be applied to a port or ports, and a frame format specifying a format applied to the corresponding rule;

receiving a frame, with the frame having multiple field values, at the port of the network device;

parsing the received frame to obtain the field values and corresponding slices;

applying the slices to the search engine;

determining matches between the slices and the common header of each rule in the rules table;

indexing into a policy table based on the determined matches to obtain an action entry; and

taking an action based on the obtained action entry.

2. The method as recited in claim 1, wherein the taking of the action based on the obtained action entry comprises at least one of dropping the frame, replacing field values in the frame, redirecting the frame, changing forwarding port map, adding forwarding port, assigning a priority of a frame, and/or changing VLAN tag.

3. The method as recited in claim 1, wherein the determining of the matches is performed by executing a parallel lookup comprising performing a parallel comparison at wire speed between the slices and each rule in the rules tables.

4. The method as recited in claim 3, further comprising:

selecting a highest priority rule when multiple matches are determined.

5. The method as recited in claim 1, wherein the determining of the matches comprises prioritizing matches obtained from the slices.

6. The method as recited in claim 1, wherein the indexing into the policy table comprises indexing into the policy table using the prioritized match as an index.

7. The method as recited in claim 1, wherein parsing of the received frame to obtain the field values comprises parsing the received frame to obtain user defined field values.

8. The method as recited in claim 1, wherein a common slice format comprises (valid, SLICE_ID, SRCPORT, SLICE_COMMON_FIELDS, slice specific fields) and the common header is the same for all slices corresponding to a same packet.

9. The method as recited in claim 1, wherein the search engine comprises a Ternary Content Addressable Memory (TCAM).

10. The method as recited in claim 1, wherein the determining of the matches between the slices and the common header of each rule in the rules table is performed in parallel.

11. The method as recited in claim 1, further comprising:

defining the source port number in the common header of the slices so that a rule applies to all ports, to only even ports, or to only a specific port.

12. The method as recited in claim 1, further comprising:

defining the frame format to comprise 1q_tagged, double tagged, 802_—

2_frame (no_tagged), cpu_frm, brcm_tagged, stack_frm, and stack_tagged.

13. The method as recited in claim 1, further comprising:

defining slices 0 to 2 of the parsed slices as predefined; and

defining slice 3 of the parsed slices as per port programmable.

14. The method as recited in claim 13, further comprising:

defining slice 0 to comprise source_port, frame_format, MACDA, MACSA, wherein MACDA is the destination MAC address and MACSA is the source MAC address.

15. The method as recited in claim 13, further comprising:

defining slice 1 to comprise source_port, frame_format tag_id, VLAN_tag, Length/Type, IP_protocol, IP_version, IP_DSCP, IP_TTL, UDFO, where TTL is a time-to-live and UDF is a user defined field.

16. The method as recited in claim 13, further comprising:

defining slice 2 to comprise source_port, IP_DA, L4_destination_port, L4_source_port, UDF1, where DA is a destination address and UDF is a user defined field.

17. The method as recited in claim 1, further comprising:

chaining multiple slices to form a single rule providing more fields than a single slice.

18. The method as recited in claim 17, further comprising:

programming the multiple slices in a chain at random locations.

19. A network device to handle a frame, comprising:

a rules table unit for defining a common header for each rule, wherein the common header comprises a source port specifying a corresponding rule to be applied to a port or ports, and a frame format specifying a format applied to the corresponding rule;

a port for receiving a frame, with the frame having multiple field values;

a pre-parser for parsing the received frame to obtain the field values and corresponding slices;

a search engine for receiving the slices and determining matches between the slices and the common header of each rule in the rules table;

a policy table unit for performing indexing based on the determined matches to obtain an action entry; and

an action resolution unit for taking an action based on the obtained action entry.

20. The network device as recited in claim 19, further comprising:

a rate meter defining a rate each frame or slice is to be processed and output.

21. The network device as recited in claim 19, wherein the action comprises at least one of dropping the frame, replacing field values in the frame, redirecting the frame, changing forwarding port map, adding forwarding port, assigning a priority of a frame, and/or changing VLAN tag.

22. The network device as recited in claim 19, wherein the search engine determines the matches by executing a parallel lookup comprising performing a parallel comparison at wire speed between the slices and each rule in the rules table unit.

23. The network device as recited in claim 22, wherein the search engine selects a highest priority rule when multiple matches are determined.

24. The network device as recited in claim 19, wherein the search engine determines the matches by prioritizing matches obtained from the slices.

25. The network device as recited in claim 19, wherein a common slice format comprises (valid, SLICE_ID, SRCPORT, SLICE_COMMON_FIELDS, slice specific fields) and the common header is the same for all slices corresponding to a same packet.

26. The network device as recited in claim 19, wherein the search engine comprises a Ternary Content Addressable Memory (TCAM).

27. The network device as recited in claim 19, wherein the search engine determines the matches between the slices and the common header of each rule in parallel.

28. The network device as recited in claim 19, wherein the source port number in the common header of the slices is defined so that a rule applies to all ports, to only even ports, or to only a specific port.

29. The network device as recited in claim 19, wherein the frame format is defined to comprise 1q_tagged, double_tagged, 802_—

2_frame (no_tagged), cpu_frm, brcm_tagged, stack_frm, and stack_tagged.

30. The network device as recited in claim 19, wherein the network device defines slices 0 to 2 of the parsed slices as predefined, and defines slice 3 of the parsed slices as per port programmable.

31. The network device as recited in claim 30, wherein slice 0 is defined to comprise source_port, frame_format, MACDA, MACSA, wherein MACDA is the destination MAC address and MACSA is the source MAC address.

32. The network device as recited in claim 30, wherein slice 1 is defined to comprise source_port, frame_format, tag_id, VLAN_tag, Length/Type, IP_protocol, IP_version, IP_DSCP, IP_TTL, UDFO, where TTL is a time-to-live and UDF is a user defined field.

33. The network device as recited in claim 30, wherein slice 2 is defined to comprise source_port, IP_DA, L4_destination_port, L4_source_port, UDF1, where DA is a destination address and UDF is a user defined field.

34. A network device to handle a frame, comprising:

rules table means for defining a common header for each rule, wherein the common header comprises a source port specifying a corresponding rule to be applied to a port or ports, and a frame format specifying a format applied to the corresponding rule;

port means for receiving a frame, with the frame having multiple field values;

pre-parser means for parsing the received frame to obtain the field values and corresponding slices;

search engine means for receiving the slices and determining matches between the slices and the common header of each rule in the rules table means;

policy table means for performing indexing based on the determined matches to obtain an action entry; and

action resolution means for taking an action based on the obtained action entry.