WO2016029684A1 - Packet processing and forwarding device and method, and computer storage medium - Google Patents

Abstract

Disclosed are a packet processing and forwarding device and method, and a computer storage medium. The device comprises processing structures. Each processing structure comprises a TCAM processing unit and a table lookup unit. The TCAM processing unit comprises: a first extraction module, used for extracting a feature field from a corresponding packet and a packet descriptor; a matching module, used for matching the feature field with TCAM items to form a matching result; and a first execution module, used for executing a primary operation on the packet and the packet descriptor according to the matching result, and forming a key value domain index. The table lookup unit comprises: a second extraction module, used for receiving the key value domain index and acquiring a key value field according to the key value domain index; a table lookup module, used for performing table lookup according to the key value field to form a table lookup result; and a second execution module, used for executing a secondary operation on the packet and the packet descriptor according to the table lookup result. The processing structure is also used for sending the packet processed by the processing structure after the specified processing structure executes the primary operation and the secondary operation.

Description

Message processing forwarding device, method and computer storage medium Technical field

The present invention relates to a packet forwarding technology in the communication field, and in particular, to a message processing and forwarding device, a method, and a computer storage medium.

Background technique

At present, packet forwarding is usually implemented by a general-purpose CPU, a fully programmable network processor NP, or a chip implemented by hard logic; a general-purpose CPU can only satisfy some extremely low line-rate requirements due to limited processing performance; Programmable network processor has high flexibility, but the cost and power consumption are relatively high. The advantages and disadvantages of hard logic implementation are also obvious. The advantage is that it can control the cost and power consumption of the chip. The disadvantage is that the chip In the development process, the understanding and combing of the business process need not be mistaken, which leads to the development and verification of the development process.

Summary of the invention

In view of this, embodiments of the present invention are directed to providing a message processing forwarding device, method, and computer storage medium to maintain the device with certain flexibility while maintaining low hardware cost and low development difficulty.

The technical solution of the embodiment of the present invention is implemented as follows:

A first aspect of the present invention provides a message processing and forwarding device, where the device includes at least one processing structure; each of the processing structures includes a TCAM processing unit and a table lookup unit;

The TCAM processing unit includes:

a first extraction module, configured to extract a feature field from the corresponding message and the message descriptor;

a matching module configured to match the feature field with a TCAM entry to form a match result;

The first execution module is a programmable module, configured to perform a first operation on the message and the message descriptor according to the matching result, and form a key value domain index;

The lookup unit includes:

a second extraction module, configured to receive the key value domain index and obtain a key value field according to the key value domain index;

The table lookup module is configured to perform a table lookup according to the key value field to form a table lookup result;

a second execution module, which is a programmable module, configured to perform a second operation on the message and the message description input according to the check result;

The processing structure is further configured to: after the specified processing structure performs the first operation and the second operation, send the message processed by the processing structure.

Based on the above scheme,

The TCAM processing unit further includes:

A peripheral logic module configured to convert a specified field in the feature field into a matching field of the TCAM entry, and the matching field is sent to the matching module.

Based on the above scheme,

The TCAM processing unit further includes:

a first acquiring module, configured to acquire operation information of the first operation according to the matching result;

The first execution module is configured to perform a first operation on the packet according to the operation information, and obtain a key field index corresponding to the matching result according to the matching result query.

Based on the above scheme,

The second extraction module is further configured to form the key value field into a predetermined key value structure and send the key value structure to the lookup table module.

Based on the above scheme,

The lookup unit includes:

a second acquiring module, configured to acquire operation information of the second operation according to the result of the table lookup;

The second execution module is configured to perform a second operation on the message according to the operation information.

Based on the above scheme,

The first execution module and the second execution module are ultra-long instruction programmable modules.

Based on the above scheme,

The TCAM processing unit and the table lookup unit are capable of processing N pieces of the feature fields at a time; the N is an integer not less than 1;

The device further includes a delay unit, and is respectively connected to the TCAM processing unit and the programmable logic instruction query unit;

The delay unit is configured to store, when the number M of the feature fields is greater than the N, a matching result corresponding to the M feature fields and the feature field, and a table lookup result, waiting for the TCAM After the processing unit and the lookup table unit process the M feature fields, the message is output.

A second aspect of the present invention provides a packet processing forwarding method, which is applied to an apparatus including at least one processing structure, where the processing structure includes a TCAM processing unit and a table lookup unit;

Extracting a feature field from the corresponding message and the message descriptor by the TCAM processing unit;

Performing a first operation on the message and the message descriptor and forming a key value domain index by using a programmable instruction in the TCAM processing unit according to the matching result;

Receiving, by the lookup table unit, the key field index, searching for a key value field according to the key field index, and performing a table lookup according to the key value field to form a table lookup result;

And using the programmable instruction in the lookup unit to use the message and the location according to the result of the lookup table The message descriptor performs a second operation;

And after the specified processing structure performs the first operation and the second operation, sending the message processed by the processing structure.

Based on the above scheme,

The TCAM processing unit includes a peripheral logic module;

The method further includes:

Converting a specified field in the feature field into a matching field of the TCAM entry by the peripheral logic module;

The matching field is used to match the TCAM entry

Based on the above scheme,

And performing, according to the matching result, performing a first operation on the message by using a programmable instruction in the TCAM processing unit and forming a key value domain index includes:

Obtaining operation information of the first operation according to the matching result;

Performing a first operation on the message according to the operation information;

The key field index corresponding to the matching result is queried according to the matching result.

Based on the above scheme,

The method further includes:

Forming the key value field into a predetermined key value structure;

The result of performing a table lookup according to the key value field is specifically:

The lookup table is formed according to the key value structure to form the lookup table result.

Based on the above scheme,

Performing the second operation on the packet by using the programmable instruction in the lookup unit according to the result of the lookup table includes:

Obtaining operation information of the second operation according to the result of the lookup table;

Performing a second operation on the message according to the operation information.

Based on the above scheme,

The programmable instructions in the TCAM processing unit and the lookup table unit are very long instructions.

Based on the above scheme,

The TCAM processing unit and the lookup table unit are capable of processing N the feature fields at a time; the N is an integer not less than 1; the device further includes a delay unit;

The method further includes: when the number M of the feature fields is greater than the N, storing, by the delay unit, the M feature fields and a matching result and a table lookup result corresponding to the feature field, Waiting for the TCAM processing unit and the lookup table unit to process the M feature fields, and outputting the message.

The third aspect of the embodiments of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, where the computer executable instructions are used to execute at least one of the methods of the first aspect of the embodiments of the present invention. one.

The message processing and forwarding method, device and computer storage medium according to the embodiment of the present invention, in combination with the TCAM processing unit and the table lookup unit, can implement the line of the device by expanding the programmable first execution module and the second execution module. Business development and business adjustment, maintaining high flexibility, compared with hard logic chips can reduce development difficulty and verification difficulty; through the introduction of TCAM processing unit, TCAM is used to partially process the message, reducing the need for programmable modules to be processed. The workload can reduce the cost and power consumption relatively.

DRAWINGS

FIG. 1 is a schematic structural diagram of a packet processing and forwarding apparatus according to an embodiment of the present invention;

2 is a schematic structural diagram of a processing structure according to an embodiment of the present invention;

3 is a schematic structural diagram of a TCAM processing unit according to an embodiment of the present invention;

4 is a second schematic structural diagram of a processing structure according to an embodiment of the present invention;

FIG. 5 is a schematic flowchart of a packet processing method according to an embodiment of the present disclosure;

FIG. 6 is a flowchart of performing a first operation and forming a key value domain index according to an embodiment of the present invention; schematic diagram;

FIG. 7 is a schematic structural view of a processing structure according to an example of the present invention.

detailed description

The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

Embodiment 1

As shown in FIG. 1 and FIG. 2, the embodiment provides a message processing and forwarding device, where the device includes at least one processing structure 11; in FIG. 1, the device includes n+1 processing structures CELL; Said n is an integer not less than one; the at least one processing structure is CELL 0, CELL 1, ... CELL n-1 and CELL n at a time;

Each of the processing structures 11 includes a TCAM processing unit 110 and a lookup unit 120;

The TCAM processing unit includes:

The first extraction module 111 is configured to extract a feature field from the corresponding message and the message descriptor;

The matching module 112 is configured to match the feature field with the TCAM entry to form a matching result;

The first execution module 113 is a programmable module, configured to perform a first operation on the message and the message descriptor according to the matching result, and form a key value domain index;

The lookup unit 120 includes:

The second extraction module 121 is configured to receive the key value domain index and obtain a key value field according to the key value domain index;

The lookup table module 122 is configured to perform a table lookup according to the key value field to form a table lookup result;

The second execution module 133 is a programmable module, configured to perform a second operation on the message and the message descriptor according to the table lookup result;

The processing structure is further configured to perform the first operation after the specified processing structure After the second operation, the message processed by the processing structure is sent.

The message is a message to be forwarded; the message descriptor is information carrying the message characteristic information; the message descriptor entering the device may include a port according to receiving the message. Configuring the obtained information and the information obtained by parsing the message; in a specific implementation process, the descriptor received by the latter processing structure further includes the processing structure formed by the previous processing structure according to the table lookup module information.

In FIG. 1, the MF_IN is a message input to the device and a message descriptor of the message; the MF_OUT is a processed output result, such as a processed message.

The TCAM is an abbreviation of Ternary Content Addressable Memory, which is a three-state content addressed memory; the TCAM entry is content stored in the TCAM. The TCAM has three states corresponding to the TCAM entry. Each bit of the TCAM entry has three states. In addition to "0" and "1", there is also a "don't care" state, so it is called " "Three-state", based on the above three states, the TCAM can perform exact matching and can perform fuzzy matching.

The TCAM processing unit is a structure including at least one of the TCAMs. In the embodiment, the TCAM processing unit preferably includes a plurality of the TCAMs, so that matching processing of multiple feature fields can be simultaneously implemented to improve Rate of forwarding processing of packets.

The first execution module is a programmable module, and the function performed by the first execution module can be changed by changing the code in the first execution module, thereby increasing the flexibility of the device.

In a specific implementation process, the matching unit may be configured to send the feature field according to the microcode to send to the matching module, where the microcode is a lower-level code, which can be easily converted from microcode to mechanical language. Can improve processing efficiency.

The packet and the packet descriptor include the MAC address, the VLAN, the MLPS, the IP address, the TTL of the lifetime label, the packet type, the packet length, the inbound port of the packet, the packet outgoing port, and the next. Information such as hop information (next network device information), priority of packet processing, and so on.

The specific structure of the first execution module 113 may be a processor, and the processor is configured to perform the first operation and form a key value domain index.

The first operation includes the first modification operation; the first modification operation may be a modification operation of a forwarding path of a message, and the like.

The first modification operation may include: obtaining, according to the matching result, a flag bit that needs to perform a specified operation; the specified operation may be a packet reporting operation, a message mirroring operation, a message sampling operation, etc.; Replace the destination MAC address in the packet, add the VLAN tag of the virtual LAN, modify the destination IP address, and modify the priority of the packet.

The first operation may also include a statistical operation; performing statistics on the forwarded packets; for example, performing statistics on the packets received from a certain port; or performing flow-based packet statistics operations.

The first operation may further include adding a discard flag or the like to the descriptor that does not satisfy the forwarding rule, for example, when detecting that the length of the IP header of the packet is less than 20 bytes, adding a discard in the descriptor. mark.

The feature field may further include packet length information, and the packet length information is mainly used to report the encapsulation time, and recalculate the length of the new packet according to the packet length and the added length of the encapsulation to ensure a new encapsulation. The length of the packet does not exceed the maximum packet length that the port can send.

In a specific process, the specific implementation of the first operation may be determined according to the TCAM entry.

The specific structure of the lookup table unit 120 may include a processor and a programmable logic device corresponding to the second execution module; the programmable logic device may perform different functions according to changes with programmable code; It has the advantages of high flexibility, low development and verification difficulty.

The specific structure of the lookup table module 122 may include a storage medium in which entry information is stored. The entry information includes a key field index and information of the second operation.

The specific structure of the second execution module 133 includes a programmable logic device, which can be modified Writing code of the programmable device to cause the second execution module to perform different functions

The second operation is similar to the first operation, and is a supplementary processing operation for the first operation, and the same includes the modification and/or determination operation of the forwarding path of the message, and the forwarding of the text. Or discard and other operations.

The device in this embodiment, in combination with the TCAM processing unit and the look-up table unit 120, uses the TCAM processing unit 110 to process conditions such as matching or selecting branches; for example, as a gateway node of a virtual local area network, the VLAN tag is parsed. The parsing of the VLAN tag may include determining the number of layers of the VLAN added in the packet. For example, whether the packet is added with only one layer of VLAN or two layers of VLANs.

The look-up table unit 120 greatly reduces the difficulty of one development due to the programmability of the code, and has the advantage of great flexibility; and because of its programmability, it leaves room for subsequent service adjustment and business development; Some basic processing is done by the TCAM processing unit, which has the advantage of being less costly than being handled entirely by the programmable module. In the embodiment, the TCAM processing unit is mainly used to complete the processing of conditional judgment and/or branch selection operation in the process of message processing and forwarding; for example, the condition is true and false.

As shown in FIG. 3, the TCAM processing unit further includes:

The peripheral logic module 114 is configured to convert the specified field in the feature field into a matching field of the TCAM entry, and the matching field is sent to the matching module.

The peripheral logic module 114 includes a circuit structure such as a logic circuit having a processing function.

The specified fields may include fields corresponding to range comparisons and/or fields requiring logical operations, and the like. Using TCAM to directly perform range comparison or logic operations will result in complex TCAM entries in the TCAM and difficulties in implementation. In the embodiment, the peripheral circuit is used to assist the execution, and not only the above functions but also the simple implementation can be realized.

Based on the foregoing solution, the TCAM processing unit further includes:

a first acquiring module, configured to acquire the operation of the first operation according to the matching result information;

The first execution module is configured to perform a first operation on the packet according to the operation information, and obtain a key field index corresponding to the matching result according to the matching result query.

The TCAM entry generally includes a field to be matched corresponding to the feature field and operation information ACTION corresponding to each matching result; if an operation such as performing VLAN resolution and/or specific IP address resolution is performed, the ACTION includes an operation corresponding to each analysis result.

In a specific implementation process, the first execution module includes a storage medium, and the storage medium has a mapping relationship between each matching result and a key value domain index; the first execution module queries the query according to the matching result. Key-valued domain index.

Based on the foregoing solution, the second extraction module is further configured to form the key value field into a predetermined key value structure, and send the key value structure to the lookup table module. In a specific implementation process, the lookup table index in the lookup table module has a fixed structure, and the key value output by the TCAM processing unit may not have a corresponding fixed structure. In this embodiment, the second extraction module is adopted. The key value field is formed into a predetermined key value structure to facilitate the lookup table of the table lookup module. The specific second extraction module is specifically configured to splicing one or more of the key value domain indexes to form a corresponding key value structure.

The lookup unit 120 includes:

a second acquiring module, configured to acquire operation information of the second operation according to the result of the table lookup;

The second execution module is configured to perform a second operation on the message and the message descriptor according to the operation information.

In a specific implementation process, the second obtaining module may include a communication interface and a processor inside the device, and the communication interface, such as a communication bus, is used to obtain operation information from the table lookup module.

The first execution module and the second execution module are both ultra-long instruction programmable units.

In the embodiment, the look-up table unit is a super-long instruction programmable logic unit, and when the at least part of the corresponding function of the table look-up unit is completed, a super-long instruction is adopted; when the device performs a long-length instruction, It is possible to realize execution of a plurality of sub-operations in the super-long instruction in parallel, which can speed up execution efficiency.

The TCAM processing unit and the table lookup unit are capable of processing N pieces of the feature fields at a time; the N is an integer not less than 1;

As shown in FIG. 4, the device further includes a delay unit 130, and is respectively connected to the TCAM processing unit and the programmable logic instruction query unit;

The delay unit 130 is configured to store, when the number M of the feature fields is greater than the N, a matching result corresponding to the M feature fields and the feature field, and a table lookup result, waiting for the After the TCAM processing unit and the lookup table unit process the M feature fields, the message is output.

The specific structure of the delay unit 130 may include a storage medium; the storage medium is configured to store the feature field and the matching result and the lookup table result.

Specifically, when the M=4; and the N=2, the TCAM processing unit and the table lookup unit cannot simultaneously perform processing of four feature fields, then the delay unit will absorb the The delays of the TCAM processing unit 110 and the lookup table unit 120 first store four feature fields, let the TCAM processing unit 110 and the lookup table unit 120 process two first, and store the two features processed first. The matching result and the query result corresponding to the field; after the feature field is processed, the first processing operation and the second operation are performed according to each feature field, and the output message is sent to the next processing structure or Forwarded by the device to the next device. In a specific implementation process, a TCAM processing unit and a table lookup unit that can simultaneously process four feature fields can also be used to implement packet processing and forwarding. Generally, the more the number of feature fields can be processed at the same time, the structure of the TCAM processing unit and the table lookup unit is complicated, and more TCAMs or processors may be needed to support concurrently processing multiple programmable instructions, such as supporting simultaneous execution. Multiple long instructions, This will result in an increase in the hardware cost of the device; in the present embodiment, the above problem can be well handled by the setting of the delay unit.

In a specific implementation process, each of the devices may be considered to correspond to one forwarding node, and after the specified processing structure is processed, the device sends the processed packet to the next forwarding node. When the message needs to pass through two or more processing structures, after the first processing structure completes the first operation and the second operation, the processed message and the message corresponding to the message are corresponding. The message descriptor is sent to the latter processing structure, and the processing structure continues to be processed by the latter processing structure. The last processing structure of the message and the message descriptor is processed, and the first operation and the second processing are processed. After the second operation, the processed message will be sent directly to the next device.

The specified processing structure may be one or more of n+1 of the processing structures.

Embodiment 2:

As shown in FIG. 5, the embodiment further provides a packet forwarding method, which is applied to an apparatus including at least one processing structure, where the processing structure includes a TCAM processing unit and a table lookup unit;

Step S110: Extract a feature field from the message and the message descriptor by using the TCAM processing unit, and match the feature field with the TCAM entry to form a matching result;

Step S120: Perform a first operation on the message and the message descriptor by using a programmable instruction in the TCAM processing unit according to the matching result, and form a key value domain index;

Step S130: extracting the key field index by using the table lookup unit, searching for a key value according to the key field index, and performing a table lookup according to the key field index to form a table lookup result;

Step S140: Perform a second operation on the message and the message descriptor by using a programmable instruction in the table lookup unit according to the result of the lookup table;

Step S150: After the specified processing structure performs the first operation and the second operation, send the message processed by the processing structure.

The method described in this embodiment may be implemented by using the apparatus as described in the first embodiment, combining the TCAM processing unit and the table lookup unit to perform the first operation and the second operation on the packet. The purpose of packet processing forwarding is achieved. For the specific structure of the TCAM processing unit and the lookup table unit, refer to the detailed description in the first embodiment.

Extracting the feature field from the description input in step S110 may be performed by the first extraction unit in the TCAM processing unit described in the first embodiment; in step S110, the feature field is matched with the TCAM entry to form a matching result, which may It is executed by the matching module of the TCAM processing unit described in the first embodiment. The step S120 can be performed by the first execution module of the TCAM processing unit described in the first embodiment.

The method in this embodiment uses the TCAM processing unit and the look-up table unit to process and forward the message, and has the advantages of using the TCAM processing unit and the programmable device, and the cost is relative to using only the programmable device. The device has the advantages of low cost and low power consumption; at the same time, it has high flexibility with respect to the curing chip, and is convenient for subsequent business adjustment and business development. In the embodiment, the TCAM processing unit mainly performs basic processing for converting condition determination or branch selection operation.

The TCAM processing unit includes a peripheral logic module;

The method also includes converting, by the peripheral logic module, a specified one of the feature fields into a match field of the TCAM entry; the match field is for matching with the TCAM entry.

In a specific implementation process, some feature fields are the same as the matching field, that is, the TCAM can also be processed. However, for the feature field with range comparison or logical operation, in the embodiment, the peripheral logic circuit of the TCAM unit can be used for conversion. The use of peripheral logic for conversion can simplify the processing of TCAM.

Optionally, as shown in FIG. 6, the step S120 includes:

Step S121: Acquire operation information of the first operation according to the matching result;

Step S122: Perform a first operation on the message and the message descriptor according to the operation information;

Step S123: Query a key field index corresponding to the matching result according to the matching result.

The first operation performed by the execution unit in this embodiment may be a modification of the information about the packet forwarding path, such as setting the upload flag of the sent message to be valid. For a detailed description, refer to the corresponding part in the first embodiment. The first operation in the embodiment of the present invention includes one or more operations performed according to the matching result. For a detailed description of the first operation, refer to the corresponding part in the first embodiment.

Optionally,

The method further includes:

Forming the key value into a predetermined key value structure;

The step S130 may specifically include: performing a lookup table according to the key value structure to form the lookup table result.

Based on the foregoing solution, the step S130 includes: acquiring operation information of the second operation according to the table lookup result; and performing a second operation on the message according to the operation information.

The result of the lookup table may be: whether the key field index matches the index in the pre-stored matching information, and specifically matches the matching information. If the key value A or the key value structure A matches the index B of the matching information, the operation information corresponding to the index B is extracted according to the matching result; and the second operation is performed according to the operation information. The operation information includes information such as a specifically performed operation, an operation type, and/or an operation parameter.

Based on the above solution, the programmable instructions in the TCAM processing unit and the look-up table unit are both ultra-long instructions; the use of ultra-long instructions has the advantage of high efficiency.

In order to simplify the structure of the TCAM processing unit and the table lookup unit and reduce the hardware cost, it is assumed that the TCAM processing unit and the lookup table unit can process N the feature fields at a time; the N is an integer not less than 1. In this embodiment, a partial feature word is post-processed. The method further includes: when the number M of the feature fields is greater than the N, storing, by the delay unit, the M feature fields and the feature field corresponding to And the result of the lookup, waiting for the TCAM processing unit and the table lookup unit to process the M feature fields, and output the message.

The embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, where the computer executable instructions are used to perform at least one of the foregoing methods, specifically as shown in FIG. 5 and/or The method shown in 6.

The computer storage medium may be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program code. In some embodiments, the computer storage medium is a non-transitory storage medium.

Two specific examples are provided below based on the first embodiment and/or the second embodiment:

Example 1:

As shown in FIG. 1, the present example provides a system including n message forwarding processing devices CELL 11; wherein n is an integer that is not less than one; and one of the processing structures CELL 11 includes a path forwarding node that can be considered to be corresponding. . The internal structure of each CELL 11 can be as shown in FIG. 7; specifically, the delay unit 200 (abbreviation MF_DELAY), the TCAM processing unit 209, and the look-up table unit 210 are included.

The TCAM processing unit 209 includes an extraction module 201 (abbreviated as T_KEY-SEL), a peripheral logic module 202 (abbreviated as LOGIC), a TCAM module 203 (abbreviated as TCAM), an acquisition module 204 (abbreviated as T_ACTION), and an execution module 205. (English abbreviation T_EX). The extraction module 201 corresponds to the first extraction module in the first embodiment; the TCAM module corresponds to the matching module in the embodiment; the acquired module and the execution module jointly correspond to the first execution in the first embodiment Module.

The lookup table unit 210 includes an extraction module 206 (English abbreviation S_KEY-SEL), an entry Module 211 (abbreviation TBL), acquisition module 208 (English abbreviation S_ACTION), and execution module 207 (English abbreviation S_EX). The extraction module 201 corresponds to the second extraction module in the first embodiment; the entry module corresponds to the look-up table module in the embodiment; the acquired module and the execution module jointly correspond to the first embodiment Second execution module.

The delay unit 200 is configured to absorb delays of item matching and table item searching, that is, to implement delay processing of part of the feature fields;

An extraction module 201, configured to extract a feature field required for matching from the description input MF-IN; the MF-OUT is an output result after processing the MF-IN;

a peripheral logic module 202, configured to improve processing capability of the TCAM module 203;

The TCAM module 203 is configured to store a matching entry converted by the microcode for matching with the feature field extracted by the module 201 and the field converted by 202; the matching entry is a TCAM entry.

The obtaining module 204 is configured to obtain ACTION information corresponding to the TCAM matching result according to the TCAM matching result;

The execution module 205 is configured to parse and execute the custom instruction corresponding to the ACTION information acquired from the module 204, and form a key value domain index.

The extracting module 206 is configured to: obtain a key field index of the lookup table according to the ACTION information obtained from the module 205, and extract the key value field into a key value structure and send the data to the table lookup unit;

An entry module 211, configured to perform a table lookup according to the key value structure;

The obtaining module 207 is configured to obtain, according to the result returned by the lookup table, the ACTION information corresponding thereto;

The execution module 208 is configured to parse and execute the custom instruction corresponding to the ACTION information acquired by the module 207.

Example two:

Based on the structure described in the first example, the forwarding process of the packet in this example is as follows:

Step 1: extract the feature field in the MF_IN from the rule written according to the microcode into the CELL unit to the TCAM processing unit 209; perform TCAM matching by the TCAM processing unit, and obtain corresponding ACTION information according to the matching result;

Step 2: Decode the obtained ACTION information, modify the message according to the decoded instruction, and generate a key field index for searching for a table;

Step 3: extracting the key value field according to the key value field index and splicing according to the agreed key value structure, and then sending it to the table item module; the table item module performs table lookup processing;

Step 4: Obtain corresponding ACTION information according to the result returned by the lookup table;

Step 5: The obtained ACTION information is decoded, and the message is modified according to the decoded instruction, and the entry address of the next CELL 11 is determined.

Step 8: After the packet is processed by multiple nodes, the real outbound port and the next hop node are determined, and the packet is forwarded.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, such as: multiple units or components may be combined, or Can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, or direct coupling, or communication connection of the components shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may be electrical, mechanical or other forms. of.

The units described above as separate components may or may not be physically separated, and the components displayed as the unit may or may not be physical units, that is, may be located in one place or distributed to multiple network units; Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated in one process. In the module, each unit may be separately used as one unit, or two or more units may be integrated into one unit; the integrated unit may be implemented in the form of hardware or a hardware plus software function unit. Formal realization.

A person skilled in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by using hardware related to the program instructions. The foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. The foregoing storage device includes the following steps: the foregoing storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. A medium that can store program code.

The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Modifications made in accordance with the principles of the invention are understood to fall within the scope of the invention.

Claims (15)

1. A message processing forwarding device, the device comprising at least one processing structure; each of the processing structures comprising a TCAM processing unit and a table lookup unit;

   The TCAM processing unit includes:

   a first extraction module, configured to extract a feature field from the corresponding message and the message descriptor;

   a matching module configured to match the feature field with a TCAM entry to form a matching result;

   The first execution module is a programmable module, configured to perform a first operation on the message and the message descriptor according to the matching result, and form a key value domain index;

   The lookup unit includes:

   a second extraction module, configured to receive the key value domain index and obtain a key value field according to the key value domain index;

   The table lookup module is configured to perform a table lookup according to the key value field to form a table lookup result;

   a second execution module, which is a programmable module, configured to perform a second operation on the message and the message description input according to the check result;

   The processing structure is further configured to: after the specified processing structure performs the first operation and the second operation, send the message processed by the processing structure.
2. The device according to claim 1, wherein

   The TCAM processing unit further includes:

   A peripheral logic module configured to convert a specified field in the feature field into a matching field of the TCAM entry, and the matching field is sent to the matching module.
3. The device according to claim 2, wherein

   The TCAM processing unit further includes:

   a first acquiring module, configured to acquire the operation of the first operation according to the matching result information;

   The first execution module is configured to perform a first operation on the packet according to the operation information, and obtain a key field index corresponding to the matching result according to the matching result query.
4. The apparatus of claim 1, wherein the second extraction module is further configured to form the key value field into a predetermined key value structure and to send the key value structure to the lookup table module.
5. The apparatus according to claim 1 or 4, wherein

   The lookup unit includes:

   a second acquiring module, configured to acquire operation information of the second operation according to the result of the table lookup;

   The second execution module is configured to perform a second operation on the message according to the operation information.
6. A device according to claim 1, 4 or 5, wherein

   The first execution module and the second execution module are ultra-long instruction programmable modules.
7. The apparatus according to any one of claims 1 to 5, wherein

   The TCAM processing unit and the table lookup unit are capable of processing N pieces of the feature fields at a time; the N is an integer not less than 1;

   The device further includes a delay unit, and is respectively connected to the TCAM processing unit and the programmable logic instruction query unit;

   The delay unit is configured to store, when the number M of the feature fields is greater than the N, a matching result corresponding to the M feature fields and the feature field, and a table lookup result, waiting for the TCAM After the processing unit and the lookup table unit process the M feature fields, the message is output.
8. A packet processing forwarding method is applied to an apparatus including at least one processing structure, where the processing structure includes a TCAM processing unit and a table lookup unit;

   Extracting feature words from corresponding message and message descriptors by the TCAM processing unit segment;

   Performing a first operation on the message and the message descriptor and forming a key value domain index by using a programmable instruction in the TCAM processing unit according to the matching result;

   Receiving, by the lookup table unit, the key field index, searching for a key value field according to the key field index, and performing a table lookup according to the key value field to form a table lookup result;

   Performing a second operation on the message and the message descriptor by using a programmable instruction in the lookup unit according to the result of the lookup table;

   And after the specified processing structure performs the first operation and the second operation, sending the message processed by the processing structure.
9. The method of claim 8 wherein

   The TCAM processing unit includes a peripheral logic module;

   The method further includes:

   Converting a specified field in the feature field into a matching field of the TCAM entry by the peripheral logic module;

   The match field is used to match the TCAM entry.
10. The method of claim 9 wherein

    And performing, according to the matching result, performing a first operation on the message by using a programmable instruction in the TCAM processing unit and forming a key value domain index includes:

    Obtaining operation information of the first operation according to the matching result;

    Performing a first operation on the message according to the operation information;

    The key field index corresponding to the matching result is queried according to the matching result.
11. The method of claim 8 wherein

    The method further includes:

    Forming the key value field into a predetermined key value structure;

    The result of performing a table lookup according to the key value field is specifically:

    The lookup table is formed according to the key value structure to form the lookup table result.
12. The method according to claim 8 or 11, wherein

    Performing the second operation on the packet by using the programmable instruction in the lookup unit according to the result of the lookup table includes:

    Obtaining operation information of the second operation according to the result of the lookup table;

    Performing a second operation on the message according to the operation information.
13. The method according to claim 8, 11 or 12, wherein

    The programmable instructions in the TCAM processing unit and the lookup table unit are very long instructions.
14. A method according to any one of claims 8 to 12, wherein

    The TCAM processing unit and the lookup table unit are capable of processing N the feature fields at a time; the N is an integer not less than 1; the device further includes a delay unit;

    The method further includes: when the number M of the feature fields is greater than the N, storing, by the delay unit, the M feature fields and a matching result and a table lookup result corresponding to the feature field, Waiting for the TCAM processing unit and the lookup table unit to process the M feature fields, and outputting the message.
15. A computer storage medium having stored therein computer executable instructions for performing at least one of the methods of claims 8-14.

Images