Parser for parsing data packets

1. A parser system comprising:

an interface configured to receive a data stream composed of interleaved sections of a plurality of different packets, and to receive section identity information about each of the sections of data defining to which packet it relates;

a scanning section configured to identify structural features of the packets using the section identify information, the structural features including the location of the start of one or more of layer 2 and layer 3 data in the packets; and

a parsing unit configured to process the data stream in a section-by-section manner and to employ the section identity information to identify and extract data from the sections of the plurality of different packets, the parsing unit comprising:

a first parser configured to extract data using offset information stored in user-programmable registers, the user offset information including an indication of whether data is to be extracted from the layer 2 or layer 3 data, and an indication of the location of the data to be extracted relative to the location of the start of the layer 2 or layer 3 data; and

a second parser configured to extract data from the packets in locations defined by the identified structural features of the packets and predetermined offset information.

2. The parser system according to claim 1 , wherein the scanning section is further configured to identify tagged packets.

3. The parser system of claim 2 wherein the scanning section is further configured to identify snapped packets.

4. The parser system according to claim 1 , wherein the interface is configured to receive a data stream including eight packets, each of which are divided into sections eight bytes in length, the sections of the eight packets being interleaved.

5. The parser system according to claim 4 , wherein the scanning section is further configured to calculate a set of variables for each eight bytes received, the set of variables including: a length variable which indicates the number of bytes received so far, an index variable which indicates the largest integer which is no greater than the length variable divided by eight, and an offset variable which indicates the length variable modulus eight.

6. The parser system according to claim 5 , wherein the scanning section is configured to identify, based on the set of variables calculated, the start of the layer 3 of the packets and whether the packets include a VLAN or SNAP tag.

7. The parser system according to claim 5 , wherein the first parser is configured to extract eight 16-bit local keys using the offset information stored in the user-programmable registers, and wherein the parsing unit is further configured to:

perform an AND operation on the extracted local keys and eight 16-bit masks;

perform an AND operation on the masks and eight 16-bit rules, and

compare the results of the AND operations to produce eight 1-bit results.

8. The parser system according to claim 7 , wherein the second parser is configured to extract a set of eight bits using the predetermined offset information provided in key selection registers, and wherein the parsing unit further comprises a combine unit configured to combine the set of eight bits and the eight 1-bit results to produce a final key.

9. A method of parsing a data stream using a computer including user-programmable registers, and a scanning section in communication with a first parser and a second parser, the method comprising:

receiving a data stream composed of interleaved sections of a plurality of different packets;

receiving section identity information about each of the sections of data defining which packet it relates to; and

processing the data stream in a section-by-section manner, said processing including:

using the scanning section to identify structural features of the packets using the section identity information, the structural features including the location of the start of one or more of layer 2 and layer 3 data in the packets

extracting, using the first parser, data from the sections of the plurality of different packets based on offset information stored in the user-programmable registers, the offset information including an indication of whether data is to be extracted from the layer 2 or layer 3 data, and an indication of the location of the data to be extracted relative to the location of the start of the layer 2 or layer 3 data; and

extracting, using the second parser, data from the sections of the plurality of different packets in locations defined by the identified structural features of the packets and predetermined offset information.

10. The method of claim 9 wherein the identifying structural features step includes identifying tagged packets.

11. The method of claim 10 wherein the identifying structural features step includes identifying snapped packets.

12. The method according to claim 9 , wherein receiving the data stream comprises receiving eight packets, each of which are divided into sections eight bytes in length, the sections of the eight packets being interleaved.

13. The method according to claim 12 , wherein identifying structural features comprises calculating a set of variables for each eight bytes received, the set of variables including: a length variable which indicates the number of bytes received so far, an index variable which indicates the largest integer which is no greater than the length variable divided by eight, and an offset variable which indicates the length variable modulus eight.

14. The method according to claim 13 , wherein identifying structural features comprises identifying, based on the set of variables calculated, the start of the layer 3 of the packets, and whether the packets include a VLAN or SNAP tag.

15. The method according to claim 12 , wherein extracting using a first parser comprises extracting eight 16-bit local keys based on the offset information stored in the user-programmable registers, the method further comprising:

performing an AND operation on the extracted local keys and eight 16-bit masks;

performing an AND operation on the masks and eight 16-bit rules, and

comparing the results of the AND operations to produce eight 1-bit results.

16. The method according to claim 15 , wherein extracting using the second parser comprises extracting a set of eight bits based on the predetermined offset information in key selection registers, the method further comprising combining the set of eight bits and the eight 1-bit results to produce a final key.