CN113364891B

CN113364891B - Network communication device and operation method of network mapping table

Info

Publication number: CN113364891B
Application number: CN202010137988.1A
Authority: CN
Inventors: 许烱发; 杨嘉瑞; 丁俊安
Original assignee: Realtek Semiconductor Corp
Current assignee: Realtek Semiconductor Corp
Priority date: 2020-03-03
Filing date: 2020-03-03
Publication date: 2023-07-18
Anticipated expiration: 2040-03-03
Also published as: CN113364891A

Abstract

The invention relates to a network communication device and a network mapping table operation method. A network communication device includes a plurality of ports, a storage medium, and a processor. The storage medium is configured to store a first lookup table and a second lookup table. The column data of the first lookup table includes a flag field. The column data of the second lookup table includes the column address of the first lookup table. Wherein the device is configured to: analyzing the packet to obtain a destination IP address; calculating column data in the second lookup table according to the destination IP address of the packet; reading a column address of the column data pointing to the column data of the first lookup table; and obtaining the destination media access control address through the content of the column data of the first lookup table, and updating the port information in real time.

Description

Network communication device and operation method of network mapping table

Technical Field

The present invention relates to a network device, and more particularly, to a network communication device operating a network address mapping table.

Background

The address resolution protocol lookup table is used for recording the corresponding relation between the network IP address and the MAC entity address. To save area when designing a network switch or router, the 48-bit MAC entity address field of the records required by the address resolution protocol table is typically shared with the MAC address field in the second layer address lookup table. It is common practice to add a field to the corresponding L2 (layer 2) table address, and the circuit can obtain the MAC entity address corresponding to the ARP entry through the L2 table. However, the MAC data recorded in the L2 table may be overwritten due to source address learning, which often results in a problem that the ARP lookup table gets a wrong MAC address. Thus, even if the switch or router reads the ARP lookup table to obtain the L2 lookup table index value, the data of the obtained L2 lookup table is already modified, resulting in an erroneous MAC entity address. Secondly, if the corresponding data item is set as a static data row to avoid data coverage, the source MAC address learning is forbidden by the data of the data row, and the updating of the source port is not executed. Another problem arises from this approach: when the host of the source MAC address has been replaced with a different port, if the L2 lookup table port field is not updated, this will result in erroneous forwarding port number information.

Therefore, the wrong forwarding port number makes the destination host not receive the packet sent from the source host for a period of time or requires software intervention to correct the wrong L2 table data, resulting in poor performance of the switch or router. Accordingly, how to solve the problem of indexing to the error data is a technical problem to be solved.

Disclosure of Invention

According to one embodiment of the present disclosure, a network communication device is disclosed, comprising: a plurality of ports configured to receive a packet; a storage medium configured to store a first lookup table and a second lookup table. A column of data of the first lookup table includes a flag field. The column data of the second lookup table contains a column address of the first lookup table. The processor is coupled to the storage medium and the ports, wherein the network communication device is configured to: the hardware or software parses the packet to obtain a source media access control address (MAC); calculating the row data in the first lookup table according to the source MAC address of the packet; reading the flag field of the column data; and determining whether the column data is referenced by the second lookup table by the contents of the flag field.

According to another embodiment, a method of operating a network mapping table is disclosed, which is applicable to a network communication device, the network communication device stores a first lookup table, a column data of the first lookup table includes a flag field, wherein the method of operating the network mapping table includes: receiving a packet from a port; parsing the packet to obtain a source Media Access Control (MAC) address; calculating the row data in the first lookup table according to the source MAC address of the packet; reading the flag field of the column data; and determining whether the column data is referenced by the second lookup table by the contents of the flag field.

Drawings

The following detailed description, when read in conjunction with the accompanying drawings, will facilitate a better understanding of the forms of the disclosure.

Fig. 1 is a schematic diagram of a network architecture in which a network communication device exchanges data with a plurality of network nodes according to some embodiments of the present disclosure.

Fig. 2 is a flowchart of a learning and updating method of a source MAC address according to some embodiments of the present application.

Fig. 3 is a flowchart illustrating a learning and updating method of a source MAC address according to other embodiments of the present disclosure.

FIG. 4 is a flow chart of querying a MAC address and mapping ports according to a destination IP address of a packet according to other embodiments of the present disclosure.

Detailed Description

Referring to fig. 1, a network architecture diagram of a network communication device 100 and a plurality of network nodes for data exchange according to some embodiments of the present disclosure is shown. The network nodes 150A-150F are presented as illustrative descriptions and are not limited to the network architecture of fig. 1. In some embodiments, the network communication device 100 is a layer two switch or a layer three router supporting an open system interconnection model.

As shown in fig. 1, the network communication device 100 includes a processor 110, a storage medium 120, and a plurality of ports 130_1, 130_2, and 130_3. The processor 110 is coupled to the storage medium 120 and a plurality of ports 130_1 to 130_3. The network nodes 150A to 150F exchange data with the ports 130_1 to 130_3 of the network communication device 100 via the local area networks lan_1 to lan_3, respectively.

In some embodiments, the network communication device 100 executes a lookup program (lookup program). For example, as shown in fig. 1, the network communication device 100 receives a packet of the network node 150A (assuming that the packet is to be transmitted to the network node 150E) through the port 130_1 belonging to the lan_1, and resolves the destination IP (destination intemet protocol, DIP) address of the packet. If the network communication device 100 obtains the destination MAC address corresponding to the destination IP address and its corresponding port (e.g., port 130_3, belonging to lan_3) in the forwarding lookup table (including the first and second lookup tables), the packet is transmitted to the network node 150E of lan_3 through the port 130_3 via the third layer route.

In some embodiments, the network communication device 100 performs a learning procedure. For example, the network node 150A generates a packet and prepares to transmit it to the network node 150E, and the network communication device 100 receives the packet through the port 130_1 and resolves the source MAC address of the packet. If the network communication device 100 queries the source MAC address in the L2 lookup table (or the first lookup table), the source port number in the packet is updated to the port mapping field of the L2 lookup table according to the setting of the column data. If the network communication device 100 cannot find the column data matching the source MAC address in the L2 lookup table, the network communication device 100 automatically adds a new column of data to the L2 lookup table, and writes the source MAC address and the source port mapping data into the column of data. In addition, whether the source MAC address is queried in the L2 lookup table or not, the survival time is reset when the row data is updated or added (refer to the following table I), which indicates that the row data is just updated.

The storage medium 120 stores first and second lookup tables. The first lookup table is a lookup table, such as an L2 lookup table, used by the layer two switch to query which port the packet should be directed to. The L2 lookup table format is shown in Table one.

Table one: l2 lookup table

In Table one, the static value indicates whether the corresponding column data is locked. In some embodiments, a static value of 1 indicates that the column data is locked, referred to as a static column; when the static value is 0, this column data is not locked, called a dynamic column. Static columns may not be updated with data content by the hardware auto-learning program, and only dynamic columns may be updated with data content by the hardware auto-learning program. The static column content can only be changed by the processor. In some embodiments, the static value field does not necessarily exist, in this case is a selective field.

In Table one, the flag indicates whether the corresponding column data is referenced by the ARP lookup table. The survival time is applied to the dynamic column to indicate whether the corresponding column data is valid, i.e., the survival time of the dynamic column data is too long to be reset, which will invalidate the column data. The port number indicates a port corresponding to the MAC address of the column data. In addition, when some dynamic column data is referenced by the ARP table, the MAC address field of the column is valid, but the survival time of the column is used to determine whether the port number data field of the column is valid.

The L2 lookup table in the prior art can only determine whether the column data is valid data according to a static value and survival time, and the action range of the column data is the whole column data; the column data contains multiple fields that are either entirely valid or entirely invalid. A flag field is added to determine whether the MAC address field and the port number field are valid by a combination of a static value, a time-to-live and a flag.

In other embodiments, the fields for static values, flags, and time to live can be generalized to function as shown in Table two below.

And (II) table: x in the table represents a value (don't care) which is not important for it

In some embodiments, when the packet belongs to internet protocol version four (IPv 4), the second lookup table is an ARP table that can be used by the network communication device 100 to look up a table of Destination IP (DIP) addresses of the third layer network protocol versus second layer Media Access Control (MAC) addresses. The ARP lookup table is shown in table three.

Table three: ARP lookup table

Unlike the normal ARP lookup table, which records the full MAC address, the column address field of the L2 lookup table in table one replaces the normal recorded MAC address. In this embodiment, the column address is address data of the L2 lookup table, and represents which column data item. For example, after the network communication device 100 obtains the destination IP address of the received packet, the ARP lookup table in table three obtains the column data corresponding to the IP address, thereby obtaining the column address of the L2 lookup table. Next, the network communication apparatus 100 reads the MAC address of table one according to the column address. Compared with the conventional ARP lookup table which needs to record the MAC address and takes 48 bits of space, the ARP lookup table of the present invention only needs 10 bits of space (when the table has 1024 columns of data), thereby achieving the advantage of space saving.

Referring again to Table one, the L2 lookup table contains fields for flags. The flag is used to indicate whether the column data in the L2 lookup table is being referenced by the ARP lookup table, and this flag will affect the updating and lookup process of the L2 lookup table.

Referring to fig. 2, a flowchart of a learning and updating method of a source MAC address according to some embodiments of the present disclosure is shown. The learning and updating method of the source MAC address is performed by the network communication device 100. This method of learning and updating the source MAC address is also referred to as the network mapping table operation method.

In step S210, a received packet is parsed to obtain the source MAC address.

In some embodiments, the data carried by the packet includes data such as destination MAC address, source MAC address, IP header, payload data, etc.

In step S220, the row address of the L2 lookup table is calculated according to the source MAC address of a received packet.

In step S230, the column data of the column address is read in the L2 lookup table.

In step S240, it is determined whether the column data is valid data.

In some embodiments, if the row is invalid, the source MAC address and source port data are written to the row and the survival time of the row data is reset in step S290. The MAC address learning procedure ends.

In some embodiments, if the row is valid data, in step S250, it is determined whether the source MAC address of the packet is the same as the MAC address of the row data.

In some embodiments, if the source MAC address of the packet is different from the MAC address of the row data, in step S300, the row address of the next row in the L2 lookup table is read. Next, the process returns to step S230 to read the column data of the next column address. When the addresses are different in step S250, in one embodiment, step S300 is performed. In another embodiment, it is also possible to end the query directly without reading down the next column address (e.g., the last column of the L2 lookup table).

In some embodiments, if the source MAC address of the packet is the same as the MAC address of the row data, in step S260, the flag of the row data is read.

In some embodiments, the flag is 1 bit of data.

In step S270, it is determined whether the column data is referenced by the ARP lookup table.

In some embodiments, when the flag is 1, it represents that this row of data is being referenced by the ARP lookup table. When the flag is 0, it represents that this column is not currently referenced by the ARP lookup table.

In some embodiments, when the content of the flag indicates that the column data is being referenced by the ARP lookup table, step S280 is performed.

In step S280, the source port number of the packet is updated to the port number field of the column data, and the survival time of the column data is updated.

In this way, if the port of the network node connected to the network communication device 100 is changed, the latest port number can be automatically learned in step S280 and automatically updated in the L2 lookup table, so that the packet can be immediately forwarded with the correct port number when the destination MAC address of the packet received next time is the same as the MAC address of the L2 lookup table.

In some embodiments, when it is determined in step S270 that the column data is not referenced by the ARP lookup table, step S310 is performed.

In step S310, a general source MAC address auto-learning procedure of the L2 lookup table is performed, which will not be described in detail.

Referring to fig. 3, a flowchart of a learning and updating method of a source MAC address according to other embodiments of the present disclosure is shown. In comparison with fig. 2, the contents of the flag are used to determine whether to update the column data of the L2 lookup table according to the received packet contents. The embodiment of fig. 3 also considers parameters such as static value, survival time, etc. of the L2 lookup table in addition to reading the contents of the flag. This method of learning and updating the source MAC address is also referred to as the network mapping table operation method.

In step S405 and step S410, a received packet is parsed to obtain a source MAC address, and a column address of the L2 lookup table is calculated according to the source MAC address.

In step S415, the network communication device 100 reads the column data of the column address from the L2 lookup table.

In some embodiments, the network communication device 100 reads the static value field of the column data.

In some embodiments, the static value is 1 bit of data. For example, when the static value of the ith column is 1, the column data representing the ith column is locked by software, so that the hardware circuit of the network communication device 100 does not dynamically update the column data. When the static value of the ith column is 0, the column data representing the column is not locked, and the hardware circuit of the network communication device 100 can update the column data through the packet data received later.

In step S420, it is determined whether the static value is equal to 1.

In some embodiments, if the static value is equal to 1, which represents that the column of data is locked, step S440 is performed.

In some embodiments, if the static value is not equal to 1, which indicates that the column data is not locked, in step S425, it is determined whether the flag of the column data is equal to 1.

In some embodiments, if the flag is equal to 1, representing that the row of data is referenced by the ARP table, step S440 is performed.

In some embodiments, if the flag is not equal to 1, which represents that the column data is not referenced by the ARP lookup table, step S430 is performed to read the survival time of the column data to determine whether the survival time is greater than the survival threshold.

In some embodiments, the survival threshold (or aging threshold) may be set to a value of 0. If the survival time of the column of data is less than or equal to the threshold (e.g., a value of 0), which represents invalid data whose contents of the port image field in the column of data are aged out, step S435 is performed.

In step S435, the source port number, source MAC address, and the reset time-to-live of the packet are overwritten with the row data.

In other embodiments, if the destination MAC address of the packet is not stored in the L2 lookup table, the network communication device 100 broadcasts the packet to all ports of the destination interface (e.g., the vpn domain) to which the packet belongs in an attempt to send the packet to the destination host. If the network communication device 100 receives a response packet, the source MAC address and the source port number of the response packet are learned into the L2 lookup table. When the same destination MAC address packet arrives at the network communication device 100 next time, it is found in the L2 lookup table and forwarded to a correct port.

In step S430, if the survival time of the row of data is greater than the threshold (e.g. a value of 0), step S440 is performed.

In step S440, if the source MAC address of the packet is the same as the MAC address of the row data, step S445 is performed. In some embodiments, when the static value or flag in the column data is 1; alternatively, if the survival time is greater than the threshold value, it can be determined that this is valid column data.

In step S445, it is again determined whether the static value is 1.

In some embodiments, if the static value is 1, step S450 is performed. In step S450, the column data is locked, so that the column data cannot be updated. The packet source MAC address learning process ends.

In some embodiments, if the static value is 0, step S460 is performed to overwrite the source port number of the packet and the reset time-to-live of the packet on the corresponding field of the column data.

In other embodiments, in step S440, if the source MAC address of the packet is different from the MAC address of the row data, step 455 is performed. Step S440 indicates that the read column data is occupied by other MAC address information.

In step 455, the column address of the next column of the L2 lookup table is read, and then returns to step 415 to continue attempting to read the next column data. In other embodiments, the network communication device 100 may stop the lookup without reading down the next column (e.g., the last column of the L2 lookup table).

In some embodiments, when the column data pointed to by the column address is occupied by other MAC addresses, the number of search steps to continue reading the next column data alignment is stopped (e.g., 4 times) after an upper limit is reached. This part is appropriately designed by those skilled in the art according to the prior art and will not be described in detail here.

It should be noted that the execution sequence of steps S420 to S430 in fig. 3 is not limited, and in other embodiments, the execution sequence of steps S425, S430, S420 may be changed, in other words, the steps may be executed according to different permutation and combination, and the same operation result may be achieved.

In some embodiments, steps S420, S445, and S450 of fig. 3 are selectively executable steps. For example, if the L2 lookup table does not have a static value field, the process is considered to be performed with a static value equal to 0, so steps S420, S445 and S450 may be omitted.

Referring to fig. 4, a flowchart of querying a MAC address and mapping ports according to a destination IP address of a packet according to some embodiments of the present disclosure is shown. This method of querying the MAC address and mapping port is also referred to as the network mapping table operation method.

In step S505, the network communication device 100 parses a received IP packet to obtain a destination IP address.

In step S510, it is determined whether the ARP table needs to be searched through a conventional IP packet searching process. This part is known to the person skilled in the art and is not described in detail here. If no ARP table lookup is required, step S545 is performed. If the ARP table needs to be looked up, step S515 is performed.

In step S545, a standard destination IP address lookup method is performed.

In step S515, a field of the ARP lookup table is calculated according to the destination IP address, and in some embodiments, the content of the field may be used to index into the L2 lookup table in the third table.

In step S520, the data in the field of the column address of the L2 lookup table in the ARP lookup table is read to obtain the column address of the L2 lookup table.

In step S525, the column data in the L2 lookup table is read according to the column address, and the static value and the field data of the survival time are obtained. Normally, the flag of the row should represent that the row of data is shared by the ARP tables, so the flag is skipped from interpretation; otherwise, the software setting error causes the two table data to be mismatched.

In steps S530 and S535, when the static value is equal to one (static column data) or the survival time is greater than the survival threshold, step S550 is performed. In some embodiments, executing step S550 indicates that the port map data is valid. If the determinations in steps S530 and S535 are negative, step S540 is performed.

In step S550, the MAC address in the column data in the L2 lookup table is read as the destination MAC address of the header of the second layer packet when the packet is transferred, and the read port data is used as the port to be transferred. At this point, the destination IP lookup procedure ends.

In step S540, the MAC address in the column data in the L2 lookup table is read as the L2 destination MAC address when the packet is transferred. Since the port field is invalid when the survival time has elapsed, the correct forwarding port cannot be determined. This packet can then only be broadcast to all VLAN ports (typically multiple ports) that are interfaced to this destination.

In some embodiments, if the MAC address of a certain column of data in the L2 lookup table is shared by the ARP table, a proposed static value is typically set to 0, except that the flag of the column is 1. The column data is represented as dynamic column data so that its port map field can be automatically updated by the packet learning procedure. In this way, even if the host changes ports, the problem that the host cannot receive the packet for a long time may not occur because the network communication device 100 does not update the port information in real time, resulting in forwarding the packet to the wrong port.

Fig. 2 to 4 illustrate the configuration of the received packet belonging to IPv 4. In other embodiments, the method for updating the network mapping table is also applicable to the architecture format of the received packet belonging to IPv 6.

In the IPv4 architecture, the network communication device 100 stores an ARP lookup table. In other embodiments, in the IPv6 architecture, the network communication device 100 stores an aryl neighbor discovery table (also called a second lookup table). The aryl neighbor discovery table contains column addresses of the L2 lookup table to correlate to column data of the L2 lookup table. Therefore, the storage medium 120 stores the L2 lookup table and the aryl neighbor discovery table, and the processes of fig. 2 to 4 can be performed as well.

In summary, the network communication device and the updating method of the network mapping table of the present invention solve the problem that when the L2 lookup table index value is used in the ARP lookup table (or the aryl neighbor discovery table), the ARP lookup table (or the aryl neighbor discovery table) gets the wrong MAC address because the L2 lookup table data is covered. In addition, the present disclosure can solve the problem that when the column data survival time field in the L2 lookup table is aged, which indicates that the port field is invalid data, the network communication device can make the destination host unable to receive the packet if the network communication device transfers the packet with reference to the port data. The field of the 1-bit flag is added in the L2 lookup table, so that the column data can be prevented from being updated by mistake under the condition that the column data is not locked, and the automatic learning of the existing L2 lookup table can be compatible. In addition, the flag judgment can be realized by designing the logic circuit through hardware, the intervention of software or a system program is not needed, and the judgment can be realized by using only 1 bit of space. Besides the reduced circuit area by using a small amount of hardware circuits, the storage space used is very small. Therefore, the technical means provided by the scheme can learn the port change in real time on the premise of saving hardware and storage space, and greatly reduce the port transferred to the error.

[ symbolic description ]

Network communication device

Processor

Storage medium

130_1, 130_2, 130_3

150A-150F

Lan_1 to lan_3

S210-S310

S405-S460

S505-S550.

Claims

1. A network communication device, comprising:

a plurality of ports configured to receive a packet;

a storage medium configured to store a first lookup table and a second lookup table, wherein a column of data of the first lookup table includes a flag field, and a column of data of the second lookup table includes a column of addresses of the first lookup table, wherein the first lookup table is a lookup table to which port the second layer switch is used to query a packet is directed, the second lookup table is an ARP table, and the flag field is used to indicate whether the column of data in the first lookup table is referenced by the second lookup table; and

a processor coupled to the storage medium and the ports, wherein the network communication device is configured to:

the hardware or software analyzes the packet to obtain a source media access control address;

calculating the column data in the first lookup table according to the source MAC address of the packet;

reading the flag field of the column data; and

it is determined whether the column data is referenced by the second lookup table by the contents of the flag field.

2. The network communication device of claim 1, wherein the column data of the first lookup table further comprises a mac address field for use in comparison with the source mac address of the packet as a destination mac address of a second layer header during packet forwarding.

3. The network communication device of claim 2, wherein the column data of the first lookup table further comprises a port number field and a time-to-live field, wherein the port number field is used for recording a source port number of the packet and is used as an output port number when the packet is forwarded for lookup, and the time-to-live field is used for indicating whether the column data is aged or not.

4. The network communication device of claim 3, wherein the column data of the first lookup table further comprises a static value field, wherein the static value field is used to represent whether the column data is locked; if the column data is locked, the column data is permanent valid data, and the column data is never updated by hardware auto-learning.

5. The network communication device of claim 4, wherein when the column data is valid data, the mac address field is the same as the source mac address of the packet and the column data is not locked, updating the source port number of the packet to the port number field of the column data, and resetting the time-to-live field of the column data when the port number field is updated.

6. The network communication device according to claim 4, wherein the source mac address and the source port number of the packet are written to the corresponding fields of the column data and the time-to-live field of the column data is reset after the mac address field is different from the source mac address of the packet, the flag field indicates that the column data is not the column data referenced by the second lookup table, and the column data is determined to be invalid.

7. The network communication device of claim 4, wherein the network communication device is configured to:

the hardware or software analyzes the packet to obtain a destination IP address;

according to the destination IP address operation of the packet, the column data of the second lookup table is obtained by searching;

obtaining the column address of the first lookup table from the column data of the second lookup table; and

the media access control address field of the column data of the first lookup table is read according to the column address.

8. The network communication device of claim 7, wherein the mac address field of the column of data is a second layer network protocol destination mac address corresponding to the destination IP address of the packet.

9. The network communication device of claim 8, wherein the output port for forwarding the packet is determined according to a port number field when the static value of the column data is in a locked state.

10. A network mapping table operation method is applicable to a network communication device, which is used for storing a first lookup table and a second lookup table, wherein a column of data of the first lookup table comprises a flag field, and a column of data of the second lookup table comprises a column of address of the first lookup table, wherein the first lookup table is a comparison table of which port a second layer switch is used for inquiring a packet is sent to, the second lookup table is an ARP table, the flag field is used for indicating whether the column of data of the first lookup table is referenced by the second lookup table, and the network mapping table operation method comprises the following steps:

receiving a packet from a port;

analyzing the packet to obtain a source MAC address;

reading the flag field of the column data; and