WO2022042543A1

WO2022042543A1 - Retransmission method for ethernet error frame and related apparatus

Info

Publication number: WO2022042543A1
Application number: PCT/CN2021/114293
Authority: WO
Inventors: 孙德胜; 牛乐宏
Original assignee: 华为技术有限公司
Priority date: 2020-08-24
Filing date: 2021-08-24
Publication date: 2022-03-03
Also published as: CN114095117A

Abstract

Disclosed in the embodiments of the present application are a retransmission method for an Ethernet error frame and a related apparatus. The method comprises: receiving an Ethernet frame; checking the Ethernet frame; when the check on the Ethernet frame fails, modifying the Ethernet frame, the modified Ethernet frame comprising a first identification which is used to indicate whether the Ethernet frame will be partially retransmitted; and then sending the modified Ethernet frame. By means of the method of the embodiments of the present application, the first identification can indicate whether an Ethernet frame in which an error occurs will be recovered by partial retransmission, and a destination device receiving the erroneous Ethernet frame can choose, according to the first identification, whether to initiate a global retransmission request, thus saving the transmission time of the error frame and improving the retransmission efficiency of the Ethernet error frame.

Description

A method and related device for retransmission of Ethernet error frame

This application claims the priority of the Chinese patent application with the application number 202010862001.2 and the application title "A method and related device for retransmission of Ethernet error frames", which was submitted to the State Intellectual Property Office of China on August 24, 2020, all of which are The contents are incorporated herein by reference.

technical field

The present application relates to the field of computer technologies, and in particular, to a method and related apparatus for retransmission of Ethernet error frames.

Background technique

The store and forward mode is a commonly used transmission mode in the Ethernet system. In this transmission method, the node device first buffers the incoming data packets from the ingress side to check the correctness of the frame (for example, based on cyclic redundancy check technology); after confirming that the packet is correct, according to the frame header The destination address carried is checked in the table to find the output port address that needs to be sent, and then the data packet is sent out. Because the node device needs to buffer the entire frame on the ingress side, the buffering delay is long, which affects the transmission efficiency.

In order to improve the transmission efficiency, those skilled in the art propose a cut-through switching (CUT-Through) method. In this transmission method, when the node device detects a data packet on the ingress side, it buffers a part of the Ethernet frame (for example, the part used for addressing), and then looks up the table according to the destination address contained in this part to find the The output port address you want to send, and then send the packet out. Compared with the store-and-forward method, the CUT-Through method can greatly reduce the transmission delay.

However, since the CUT-Through method does not cache the complete data packet, in the case of an error in the data packet, the node device cannot discard the erroneous data packet after detecting the data packet error (because the data packet starts to search after a part of the data packet is buffered. address forwarded). Therefore, in the case of using the CUT-Through method, in order to ensure the reliability of the data, it is necessary to rely on the verification mechanism of the upper layer of the receiving end to verify the received data packets, such as cyclic redundancy check (CRC) . When the receiver detects that it has received an erroneous packet, it needs to send a retransmission request to the initiator. Then, the transmission delay of this packet will be 3 times the transmission delay when no error occurs. How to improve the retransmission efficiency of Ethernet error frames is an urgent problem to be solved by those skilled in the art.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a method and a related device for retransmission of Ethernet error frames, which can improve the retransmission efficiency of Ethernet error frames.

In a first aspect, an embodiment of the present application provides a method for retransmitting an Ethernet error frame, the method comprising: receiving an Ethernet frame; verifying the Ethernet frame; when the Ethernet frame verification fails , modifying the Ethernet frame, where the modified Ethernet frame includes a first identifier, where the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted; and the modified Ethernet frame is sent. In this way, the first identifier can indicate whether the erroneous Ethernet frame will be recovered by local retransmission, and the destination device receiving the erroneous Ethernet frame can choose whether to initiate a global retransmission request according to the first identifier, which can save The transmission time of error frames improves the retransmission efficiency of Ethernet error frames.

With reference to the first aspect, in a possible implementation manner, the modified Ethernet frame further includes a second identifier, where the second identifier is used to indicate that the modified Ethernet frame is invalid. In this implementation manner, whether the Ethernet frame is valid can be determined through the second identifier.

With reference to the first aspect, in a possible implementation manner, if the Ethernet frame is to be partially retransmitted, the retransmitted Ethernet frame is obtained; the retransmitted Ethernet frame is checked; When the retransmitted Ethernet frame is successfully verified, the retransmitted Ethernet frame is sent. In this implementation manner, in the case of detecting that an error occurs in the Ethernet frame, the node device in the transmission process can obtain the retransmitted Ethernet frame. Then, the destination device receiving the Ethernet frame does not need to send a global retransmission request to the sending device of the Ethernet frame, which saves the transmission time of the error frame and improves the retransmission efficiency of the Ethernet error frame.

With reference to the first aspect, in a possible implementation manner, the obtaining the retransmitted Ethernet frame includes: sending a partial retransmission request for the Ethernet frame; receiving a partial retransmission request sent in response to the partial retransmission request Retransmitted Ethernet frame.

With reference to the first aspect, in a possible implementation manner, the obtaining the retransmitted Ethernet frame includes: performing recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame.

With reference to the first aspect, in a possible implementation manner, if it is detected that the Ethernet frame is invalid, the Ethernet frame is sent. In this implementation manner, if the Ethernet frame check fails and is invalid, it indicates that the Ethernet frame has been modified; then the Ethernet frame does not need to be modified this time.

With reference to the first aspect, in a possible implementation manner, before the modification of the Ethernet frame, the method further includes: determining whether the Ethernet frame has been modified, and if not, performing modification The step of the Ethernet frame; if it has been modified, the Ethernet frame is not modified, and the Ethernet frame is directly sent.

With reference to the first aspect, in a possible implementation manner, the first identifier is located after the payload field of the Ethernet frame. In this implementation manner, since the Ethernet frames are sent in sequence according to the order of byte arrangement, the manner in which the first identifier is located after the payload field can facilitate the modification of the Ethernet frame.

With reference to the first aspect, in a possible implementation manner, the checking the Ethernet frame includes: checking the Ethernet frame and sending some bits in the Ethernet frame; the Sending the modified Ethernet frame includes: continuing to send bits in the modified Ethernet frame after the partial bits are removed.

In a second aspect, the embodiments of the present application provide another method for retransmitting an Ethernet error frame, the method includes: receiving an Ethernet frame, and verifying the Ethernet frame; when the Ethernet frame verification is successful If the Ethernet frame is invalid, check whether the Ethernet frame will be partially retransmitted; when it is detected that the Ethernet frame will be partially retransmitted, wait for receiving Retransmitted Ethernet frame. In this way, after receiving an erroneous Ethernet frame, if it is detected that the erroneous Ethernet frame will be partially retransmitted, the destination device waits to receive the retransmitted Ethernet frame, and does not need to send any further messages to the Ethernet frame. The sending device sends a global retransmission request, which can save the transmission time of error frames and improve the retransmission efficiency of Ethernet error frames.

With reference to the second aspect, in a possible implementation manner, the second identifier is used to indicate that the Ethernet frame is invalid, and the detecting whether the Ethernet frame is valid includes: detecting whether the Ethernet frame contains the second If it is included, it is determined that the Ethernet frame is invalid; if it is not included, it is determined that the Ethernet frame is valid. In this implementation manner, whether the Ethernet frame is valid can be determined by using the second identifier.

With reference to the second aspect, in a possible implementation manner, the Ethernet frame includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid, and the detecting whether the Ethernet frame is valid , including: determining whether the Ethernet frame is valid according to the second identifier; if the second identifier is in the first state, determining that the Ethernet frame is invalid; if the second identifier is in the second state, then It is determined that the Ethernet frame is valid. In this implementation manner, whether the Ethernet frame is valid can be determined by using the second identifier.

With reference to the second aspect, in a possible implementation manner, the Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame is to be partially retransmitted, and the detecting the Ethernet frame Whether the network frame will be partially retransmitted includes: determining whether the Ethernet frame will be partially retransmitted according to the first identifier; if the first identifier is in the fourth state, determining that the Ethernet frame will be partially retransmitted transmission; if the first identifier is in the fifth state, it is determined that the Ethernet frame will not be partially retransmitted. In this implementation manner, whether the Ethernet frame will be partially retransmitted may be determined by the first identifier.

With reference to the second aspect, in a possible implementation manner, the method further includes: when it is detected that the Ethernet frame will not be locally retransmitted, sending a global retransmission request for the Ethernet frame.

With reference to the second aspect, in a possible implementation manner, the method further includes: when the Ethernet frame check fails, determining whether the Ethernet frame can be recovered locally; when determining whether the Ethernet frame can be recovered locally; Local recovery is performed on the Ethernet frame when it can be recovered locally.

With reference to the second aspect, in a possible implementation manner, the method further includes: when it is detected that the Ethernet frame cannot be recovered locally, sending a global retransmission request for the Ethernet frame.

With reference to the second aspect, in a possible implementation manner, the method further includes: when the Ethernet frame check fails, determining whether the Ethernet frame can be recovered locally, including: when the Ethernet frame fails When the frame check fails, it is detected whether the Ethernet frame is valid; if the Ethernet frame is valid, it is determined whether the Ethernet frame can be recovered locally. In this implementation manner, the Ethernet frame check fails and is valid, indicating that an error occurs in the transmission of the Ethernet frame on the Ethernet link between the destination device and the previous node device, and the Ethernet frame is in the Ethernet link. If no errors occurred during the previous transmission, the destination device attempts to recover the Ethernet frame. Therefore, the destination device does not need to send a global retransmission request to the sending device of the Ethernet frame, which saves the transmission time of the erroneous frame and improves the retransmission efficiency of the Ethernet erroneous frame.

With reference to the second aspect, in a possible implementation manner, the method further includes: if the Ethernet frame is invalid, waiting to receive a retransmitted Ethernet frame. The Ethernet frame check fails and is invalid, indicating that the Ethernet frame has an error in the transmission of the Ethernet link between the destination device and the previous node device, and the Ethernet frame has been transmitted before the Ethernet link. If an error occurs and the Ethernet frame has been modified, the destination device can wait for the retransmitted Ethernet frame obtained by the receiving node device. Therefore, the destination device does not need to send a global retransmission request to the sending device of the Ethernet frame, which saves the transmission time of the erroneous frame and improves the retransmission efficiency of the Ethernet erroneous frame.

With reference to the second aspect, in a possible implementation manner, after the waiting for receiving the retransmitted Ethernet frame, the method further includes: if the current time is more than a preset time period from the time when the Ethernet frame is received , a global retransmission request for the Ethernet frame is sent. In this implementation manner, in order to ensure that the correct Ethernet frame can be obtained, the destination device will send a global retransmission request for the Ethernet frame when it is detected that the waiting duration exceeds the preset duration.

In a third aspect, an embodiment of the present application provides a communication device, the communication device includes a receiving unit, a checking unit, a modifying unit, and a sending unit, wherein: the receiving unit is configured to receive an Ethernet frame; The verification unit is used to verify the Ethernet frame; the modification unit is used to modify the Ethernet frame when the Ethernet frame verification fails, and the modified Ethernet frame includes the first identifier , the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted; the sending unit is used to send the modified Ethernet frame.

With reference to the third aspect, in a possible implementation manner, the modified Ethernet frame further includes the second identifier, where the second identifier is used to indicate that the modified Ethernet frame is invalid.

With reference to the third aspect, in a possible implementation manner, the communication device further includes an acquisition unit, the acquisition unit is configured to acquire the retransmitted Ethernet frame in the case that the Ethernet frame will be partially retransmitted network frame; the verification unit is further configured to verify the retransmitted Ethernet frame; the sending unit is further configured to send the retransmitted Ethernet frame when the verification of the retransmitted Ethernet frame is successful Retransmitted Ethernet frame.

With reference to the third aspect, in a possible implementation manner, the obtaining unit is specifically configured to: send a partial retransmission request for the Ethernet frame; receive a retransmitted Ethernet frame sent in response to the partial retransmission request web frame.

With reference to the third aspect, in a possible implementation manner, the obtaining unit is specifically configured to: perform recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame.

With reference to the third aspect, in a possible implementation manner, the first identifier is located after the payload field of the Ethernet frame.

With reference to the third aspect, in a possible implementation manner, the check unit is specifically configured to: check the Ethernet frame and send some bits in the Ethernet frame; the sending unit is specifically configured to: To: continue to send bits in the modified Ethernet frame after the partial bits are removed.

In a fourth aspect, an embodiment of the present application provides a communication device, the communication device includes a verification unit, a first detection unit, a second detection unit, and a receiving unit, wherein: the verification unit is configured to receive Ethernet frame, to verify the Ethernet frame; the first detection unit is used to detect whether the Ethernet frame is valid when the Ethernet frame verification is successful; the second detection unit is used to If the Ethernet frame is invalid, detect whether the Ethernet frame will be partially retransmitted; the receiving unit is configured to wait for receiving the retransmitted Ethernet frame when it is detected that the Ethernet frame will be partially retransmitted web frame.

With reference to the fourth aspect, in a possible implementation manner, the second identifier is used to indicate that the Ethernet frame is invalid, and the first detection unit is specifically configured to: detect whether the Ethernet frame contains the second identifier; if If it is included, it is determined that the Ethernet frame is invalid; if it is not included, it is determined that the Ethernet frame is valid.

With reference to the fourth aspect, in a possible implementation manner, the Ethernet frame includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid, and the first detection unit is specifically configured to: Determine whether the Ethernet frame is valid according to the second identifier; if the second identifier is in the first state, determine that the Ethernet frame is invalid; if the second identifier is in the second state, determine the Ethernet frames are valid.

With reference to the fourth aspect, in a possible implementation manner, the Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted, and the second detection unit It is specifically used to: determine whether the Ethernet frame will be partially retransmitted according to the first identifier; if the first identifier is in the fourth state, determine that the Ethernet frame will be partially retransmitted; If it is identified as the fifth state, it is determined that the Ethernet frame will not be partially retransmitted.

With reference to the fourth aspect, in a possible implementation manner, the communication apparatus further includes a first sending unit, where the first sending unit is configured to: when it is detected that the Ethernet frame will not be partially retransmitted, A global retransmission request for the Ethernet frame is sent.

With reference to the fourth aspect, in a possible implementation manner, the communication device further includes a determination unit and a recovery unit, and the determination unit is configured to determine the Ethernet frame when the Ethernet frame check fails Whether it can be recovered locally; the recovery unit is configured to perform local recovery on the Ethernet frame when it is determined that the Ethernet frame can be recovered locally.

With reference to the fourth aspect, in a possible implementation manner, the determining unit is specifically configured to: when the Ethernet frame check fails, detect whether the Ethernet frame is valid; if the Ethernet frame is valid, It is then determined whether the Ethernet frame can be recovered locally.

With reference to the fourth aspect, in a possible implementation manner, the receiving unit is further configured to: if the Ethernet frame is invalid, wait to receive the retransmitted Ethernet frame.

With reference to the fourth aspect, in a possible implementation manner, the communication device further includes a second sending unit, and the second sending unit is configured to: if the current time is more than a preset time from the time when the Ethernet frame is received duration, then send a global retransmission request for the Ethernet frame.

In a fifth aspect, the present application provides yet another communication device, the communication device includes a processor, a memory, and a transceiver; the transceiver is used to receive Ethernet frames; the memory is used to store program codes; The processor is configured to call the program code from the memory to execute the method described in the first aspect or any possible implementation manner of the first aspect.

In a sixth aspect, the present application provides yet another communication device, the communication device includes a processor, a memory, and a transceiver; the transceiver is used to receive Ethernet frames; the memory is used to store program codes; The processor is configured to call the program code from the memory to execute the method described in the second aspect or any possible implementation manner of the second aspect.

In a seventh aspect, the present application provides a computer-readable storage medium, where the computer-readable storage medium is used to store instructions, and when the instructions are executed, the first aspect or any one of the first aspect may be possible. The method described by the implementation is implemented.

In an eighth aspect, the present application provides another computer-readable storage medium, where the computer-readable storage medium is used to store instructions, and when the instructions are executed, make the second aspect or any one of the second aspects described above. Possible implementations The described method is implemented.

In a ninth aspect, the present application provides a chip system, the chip system includes at least one processor and an interface for supporting the first device to implement the functions involved in the first aspect, for example, receiving or processing the functions involved in the above method. at least one of the data and information. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the site. The chip system may be composed of chips, or may include chips and other discrete devices.

In a tenth aspect, the present application provides another chip system, the chip system includes at least one processor and an interface for supporting the first device to implement the functions involved in the second aspect, for example, receiving or processing the functions involved in the above method. At least one of the data and information involved. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the access point. The chip system may be composed of chips, or may include chips and other discrete devices.

In this embodiment of the present application, the first identifier may indicate whether the erroneous Ethernet frame will be recovered by local retransmission, and the destination device that receives the erroneous Ethernet frame may select whether to initiate a global retransmission request according to the first identifier, It can save the transmission time of error frames and improve the retransmission efficiency of Ethernet error frames.

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that are required to be used in the description of the embodiments or the prior art.

1 is a schematic diagram of a frame format of an Ethernet frame provided by an embodiment of the present application;

2 is a schematic diagram of a system architecture of an Ethernet system provided by an embodiment of the present application;

3 is a schematic diagram of a retransmission process of an Ethernet error frame provided by an embodiment of the present application;

4 is a schematic diagram of a retransmission process of another Ethernet error frame provided by an embodiment of the present application;

5 is a flowchart of a method for retransmitting an Ethernet error frame provided by an embodiment of the present application;

6 is a schematic diagram of a frame format of another Ethernet frame provided by an embodiment of the present application;

7 is a schematic diagram of a frame format of another Ethernet frame provided by an embodiment of the present application;

8A is a flowchart of another method for retransmitting an Ethernet error frame provided by an embodiment of the present application;

FIG. 8B is a flowchart of another method for retransmitting an Ethernet error frame provided by an embodiment of the present application

9A is a schematic diagram of a retransmission process of another Ethernet error frame provided by an embodiment of the present application;

9B is a schematic diagram of a retransmission process of another Ethernet error frame provided by an embodiment of the present application;

10 is a schematic diagram of a retransmission process of another Ethernet error frame provided by an embodiment of the present application;

11A is a flowchart of another method for retransmitting an Ethernet error frame provided by an embodiment of the present application;

11B is a flowchart of another method for retransmitting an Ethernet error frame provided by an embodiment of the present application;

FIG. 12 is a schematic diagram of a communication device provided by an embodiment of the present application;

FIG. 13 is a schematic diagram of another communication device provided by an embodiment of the present application;

FIG. 14 is a schematic diagram of another communication device provided by an embodiment of the present application;

FIG. 15 is a schematic structural diagram of a chip provided by an embodiment of the present application.

detailed description

The technical solutions in the embodiments of the present application are described in more detail below.

The technical solution of the present application can be applied to Ethernet (Ethernet), which is a computer local area network. For example, the application scenario of the present application may be an Ethernet based on the Institute of Electrical and Electronics Engineers (IEEE) 802.3 standard or the IEEE 802.1 standard, or an Ethernet based on the IEEE 802.3 next-generation standard.

Some concepts involved in the embodiments of the present application are introduced below.

1. Ethernet frame

Ethernet frames are packets of data transmitted over an Ethernet link. Referring to FIG. 1, FIG. 1 is a schematic diagram of a frame format of an Ethernet frame provided by an embodiment of the present application. The Ethernet frame can contain the following fields:

The preamble field, including 7 bytes, alternates between 1 and 0 of the preamble. The adapter at the receiving end can adjust the clock frequency according to the preamble field when receiving a media access control (media access control, MAC) frame, so that it is synchronized with the frequency of the sending end. The frame start delimiter (SFD) field includes 1 byte, in which the first six bits of the 8 bits are alternated with 1 and 0, and the last two bits are continuous 1, indicating that the subsequent content is frame information. The destination address field, including 6 bytes, is used to indicate the media access control (media access control, MAC) address of the receiving device of the Ethernet frame. The source address field, including 6 bytes, is used to indicate the MAC address of the sending device of the Ethernet frame. The type field, including 2 bytes, is used to mark the protocol used by the upper layer, so that the payload field of the received MAC frame can be handed over to the protocol of the upper layer. For example, when the field is 0x0800, it means that the payload (or payload, data) is delivered to the Internet Protocol (IP); when the field is 0x0806, it means that the payload is delivered to the Address Resolution Protocol (Address). Resolution Protocol, ARP); when this field is 0X8035, it indicates that the payload is delivered to the Reverse Address Resolution Protocol (RARP). In some standards, the 2 bytes where the type field is located is the length field, which is used to indicate the length of the payload field. Exemplarily, the field can be determined according to the numerical value of this field. If it is less than or equal to 1536 (0x600), this field is a length field; if it is greater than 1536, this field is a type field. The payload field, also known as the payload field or the data field, includes the data that needs to be delivered to the upper layer. Among them, the minimum length of Ethernet frame data is 46 bytes, and the maximum is 1500 bytes (or 1504, 1982 bytes, different standards can have different regulations), if the data is less than 46 bytes, it will use Padding bytes are padded to the minimum length. The maximum value is also called the maximum transmission unit (MTU). The Frame Check Sequence (FCS) field, consisting of 4 bytes, is used to detect whether an error has occurred in the Ethernet frame. In some embodiments, the Ethernet frame may also include an extension field.

2. Ethernet frame check

The node device verifies the received Ethernet frame. If the verification is successful, it indicates that there is no error in the transmission of the Ethernet frame between the node device and the previous node device; if the verification fails , indicating that the Ethernet frame has an error during the transmission of the Ethernet link between the node device and the previous node device.

In some implementations, the Ethernet frame can be checked by means of CRC check.

The principle of CRC check is: the sending device calculates the CRC value according to the preset calculation method according to the data to be sent, and sends it to the receiving device together with the data. Exemplarily, referring to FIG. 1 , the FCS field contains a CRC value. The receiving device calculates the CRC value according to the same calculation method for the received data, and then the receiving device compares the received CRC value with the calculated CRC value. Specifically, if the two CRC values are different, the Ethernet frame check fails; if the two CRC values are the same, the Ethernet frame check succeeds.

However, it should be noted that the node device cannot know whether an error occurred in the previous transmission process of the Ethernet frame through the check result. For example, if the Ethernet frame sent by the previous node device to the node device is originally a wrong Ethernet frame, and the wrong Ethernet frame is in the transmission process of the Ethernet link from the previous node device to the node device. If no error occurs, then the verification of the Ethernet frame by the node device is also successful.

In addition, the node device may also use other check methods for the Ethernet frame, for example, check sum (check sum), bit interleaved parity (bit interleaved parity, BIP), etc., and the embodiment of the present application does not limit the checksum The way. The following embodiments mainly take the CRC check as an example for introduction.

3. Ethernet frame is valid

Since the verification result of the Ethernet frame can only indicate whether an error occurs during the transmission of the Ethernet frame between the node device and the previous node device, after the destination device successfully checks the received Ethernet frame , it is also necessary to judge the validity of the Ethernet frame. The destination device is the device that finally receives the Ethernet frame, and the MAC address of the destination device is the same as the MAC address contained in the destination address field in the Ethernet frame.

Specifically, if the Ethernet frame is valid, it means that there is no error in the transmission process of the Ethernet frame before the Ethernet link between the destination device and the previous node device; if the Ethernet frame is invalid, it means that the Ethernet frame is between the destination device and the previous node device. An error occurred during the previous transmission of the Ethernet link of the previous node device. In the following content, the method for determining whether the Ethernet frame is valid will be introduced in detail.

4. Local retransmission

In this embodiment of the present application, if the node device fails to verify the Ethernet frame, the node device can obtain the retransmitted Ethernet frame by means of partial retransmission. The meaning of local retransmission is: the node device recovers the Ethernet frame through its own capabilities, and obtains the retransmitted Ethernet frame; The Ethernet link with the previous node device obtains the retransmitted Ethernet frame from the previous node device. The retransmission request for the Ethernet frame sent by the node device to the previous node device may be referred to as a partial retransmission request. In addition, after acquiring the retransmitted Ethernet frame, the node device sends the retransmitted Ethernet frame to the next node device.

5. Global retransmission

In the embodiment of the present application, the meaning of global retransmission is: the destination device sends a retransmission request for the Ethernet frame to the source device, and obtains the retransmission from the source device through the Ethernet link between the destination device and the source device. Ethernet frame. The source device is a device that generates the Ethernet frame, and the MAC address of the source device is the same as the MAC address included in the source address field of the Ethernet frame. The retransmission request for the Ethernet frame sent by the destination device to the source device may be called a global retransmission request.

6. Local recovery

In this embodiment of the present application, after the destination device fails to verify the received Ethernet frame, the recovered Ethernet frame may be acquired in a local recovery manner. The meaning of local recovery is: the destination device recovers the Ethernet frame through its own capabilities, and obtains the recovered Ethernet frame; An Ethernet link between node devices obtains recovered Ethernet frames from the previous node device. After acquiring the recovered Ethernet frame, the destination device verifies the recovered Ethernet frame again, and for the subsequent operations, please refer to the introduction in the subsequent content.

Referring to FIG. 2 , it is a schematic diagram of a system architecture of an Ethernet system provided by an embodiment of the present application. As shown in FIG. 2 , the Ethernet system includes computer equipment (for example, a first computer equipment, a second computer equipment) and switches (for example, a first switch, a second switch, and a third switch). In this Ethernet system, node devices are connected by Ethernet links to realize data transmission. It should be noted that the Ethernet system includes at least two node devices, and may include more or less devices than those shown in FIG. 2 , which is not limited here, and only FIG. 2 is used as an example. in:

Computer equipment, which can be used to realize functions such as data generation, reception, data transmission, and data processing. Optionally, the computer device may also be a network device such as a server (for example, a network management server, a device management server, an authentication server, and a positioning server).

The switch can be used to realize functions such as data transmission. It can send the data to be transmitted to the destination address according to the information transmission requirements of both ends of the communication. Optionally, the switch may also be a network connection device such as a network card, a hub, and a router.

Exemplarily, the first computer device generates an Ethernet frame and sends the Ethernet frame to the second computer device. During the transmission of the Ethernet frame, through the Ethernet links "first computer device-first switch", "first switch-second switch", "second switch-third switch" and "third switch-first switch" Two Computer Devices", and Ethernet Node Devices "First Switch", "Second Switch" and "Third Switch". The method for retransmitting an Ethernet error frame provided by the embodiment of the present application is applicable to the transmission process of the Ethernet frame in the Ethernet system.

First, some methods for retransmitting Ethernet error frames in the prior art are introduced, taking the first computer device generating the Ethernet frame and sending the Ethernet frame to the second computer device as an example.

In one approach, Ethernet transmits Ethernet frames in a store-and-forward manner. In this transmission mode, the node device first buffers the incoming data packets from the ingress side, and checks the correctness of the frame (for example, checking based on the cyclic redundancy check technology). If the detected Ethernet frame is correct, the node device will look up the table according to the destination address carried in the frame header, find the output port address to be sent, and then send the Ethernet frame. If an error is detected in the Ethernet frame, the node device discards the Ethernet frame.

Referring to FIG. 3 , it is a schematic diagram of a retransmission process of an Ethernet error frame provided by an embodiment of the present application. For an Ethernet frame, H is a frame header (head) part, P is a payload (or called a payload) (Payload) part, and F is a frame check sequence (frame check sequence, FCS) part. It should be noted that the examples in the figure in part F are F ₁ and F ₂ , in order to distinguish two different frame check sequences. Exemplarily, the F part may include a CRC-32 field for checking the correctness of the Ethernet frame. Among them, two cases of correct transmission of Ethernet frames and errors in transmission are illustrated.

The case of correct transmission: the first switch receives the Ethernet frame sent by the first computer device, and the first switch performs CRC check on the Ethernet frame. In the case of correct transmission, the CRC value calculated by the receiving device (F ₁ in the example) is the same as the CRC value contained in the received Ethernet frame (F ₁ in the example).

After the first switch successfully checks the Ethernet frame, it sends the Ethernet frame to the second switch. According to a similar process, the Ethernet frame is transmitted through the second switch and the third switch, and finally reaches the destination receiving device, that is, the second computer device.

In the case of an error in transmission (taking the Ethernet frame error code on the Ethernet link "the first computer device - the first switch" as an example): the first switch receives the Ethernet frame sent by the first computer device, and the first switch pairs Ethernet frames are CRC checked. Due to an error occurred during the transmission of the Ethernet frame, the CRC value calculated by the first switch (F ₂ in the example) is different from the received CRC value (F ₁ in the example), then it is determined that the Ethernet frame is in error, then Drop the Ethernet frame. The second computer device detects the packet loss (for example, uses a timeout mechanism to determine the packet loss and determines that the next Ethernet frame is received), and then sends a retransmission request (for example, Negative Acknowledgment, NAK) to the first computer device. Arts). Next, the first computer device sends the retransmitted Ethernet frame to the second computer device according to the retransmission request. It can be seen that, by adopting this method of retransmission of the erroneous frame, the delay in receiving the retransmitted Ethernet frame will be at least three times that of the case where no error occurs in the transmission of the Ethernet frame.

In another way, in order to improve the transmission efficiency, those skilled in the art propose a transmission way of cut-through switching (CUT-Through). In this transmission mode, when the node device detects a data packet on the ingress side, it buffers a part of the Ethernet frame (for example, the part used for addressing, such as the frame header), and then according to the destination address contained in this part Look up the table, find the output port address you want to send, and then send the packet. Compared with the store-and-forward method, the CUT-Through method can greatly reduce the transmission delay. However, since the CUT-Through method does not cache the complete data packet, in the case of an error in the data packet, the node device cannot discard the erroneous data packet after detecting the data packet error (because the data packet starts to search after a part of the data packet is buffered. address forwarded). Therefore, in the case of using the CUT-Through method, in order to ensure the reliability of the data, it is necessary to rely on the verification mechanism of the upper layer of the receiving end (the upper layer of the MAC layer, that is, the IP layer, the application layer, etc.) to verify the received data packets. Check, such as cyclic redundancy check (cyclic redundancy check, CRC). When the receiver detects that it has received an erroneous packet, it needs to send a retransmission request to the initiator.

Referring to FIG. 4 , it is a schematic diagram of another retransmission process of an Ethernet error frame provided by an embodiment of the present application. The Ethernet system architecture of FIG. 4 is similar to that of FIG. 2. The first computer device generates an Ethernet frame and sends the Ethernet frame to the second computer device. For the transmission process of the Ethernet, reference may be made to the above introduction, and details are not repeated here. In the Ethernet frame, H is the head part, P is the payload part, F is the FCS part, and f is the check part (for example, it can be a CRC-16 field). The second computer device may, according to f, rely on the application layer verification mechanism to determine whether the received Ethernet frame is in error.

The case of correct transmission: the first switch receives the Ethernet frame sent by the first computer device, and the first switch performs CRC check on the Ethernet frame. After the first switch successfully checks the Ethernet frame, it sends the Ethernet frame to the second switch. According to a similar process, the Ethernet frame is transmitted through the second switch and the third switch, and finally reaches the destination receiving device, that is, the second computer device. The second computer device may, according to f, rely on the application layer verification mechanism to determine that the received Ethernet frame is correct.

When there is an error in transmission (take the Ethernet frame error code in the Ethernet link "the first computer device - the first switch" as an example): the first switch first buffers the head part of the Ethernet frame, and then according to the purpose contained in the head part Check the address table, find the output port address to be sent, and then send the Ethernet frame to the second switch. An error occurs in the f part due to a bit error, and an example of this part after the error occurs is the f' part. The first switch sends the erroneous Ethernet frame to the second switch. Since the Ethernet frame received by the second switch is already an Ethernet frame with errors, the CRC value of the Ethernet frame received by the second switch (F ₂ in the example) is the same as the calculated CRC value (F ₂ in the example) , the second switch successfully checks the wrong Ethernet frame. Afterwards, the second switch sends the errored Ethernet frame to the third switch, and the third switch sends the errored Ethernet frame to the second computer device. The second computer device verifies the received Ethernet frame successfully, and then the second computer device determines that the received Ethernet frame is an erroneous Ethernet frame by relying on the upper-layer verification mechanism according to f'. Next, the second computer device sends a retransmission request to the first computer device, and correspondingly, the first computer device sends the retransmitted Ethernet frame to the second computer device according to the retransmission request. It can be seen that, by adopting the retransmission method of the erroneous frame, the delay in receiving the retransmitted Ethernet frame will be at least three times that of the case where no error occurs in the transmission of the Ethernet frame.

In view of this, a method for retransmitting an Ethernet error frame according to an embodiment of the present application is provided. In the embodiment of the present application, during the transmission of the Ethernet frame, a node device checks the Ethernet frame after receiving the Ethernet frame sent by the previous node device; when the Ethernet frame check fails When the node device modifies the Ethernet frame, the modified Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted; next, the node device retransmits The modified Ethernet frame is sent to the next node. In this way, the first identification can indicate whether the Ethernet frame in which the error occurred will be recovered by partial retransmission. The destination device receiving the erroneous Ethernet frame can select whether to initiate a global retransmission request according to the first identifier, which can save the transmission time of the erroneous frame and improve the retransmission efficiency of the Ethernet erroneous frame.

Referring to FIG. 5, FIG. 5 is a flowchart of a method for retransmitting an Ethernet error frame provided by an embodiment of the present application. The method may be implemented based on the Ethernet system shown in FIG. 1 , and the second node device described below may be the first switch, the second switch or the third switch shown in FIG. 1 . The first node device is the previous node device of the second node device, and the third node device is the next node device of the second node device. The previous node device and the next node device are determined according to the transmission direction of the Ethernet frame. Exemplarily, in an embodiment, the second node device is a first switch, then the first node device is a first computer device, and the third node device is a second switch. In yet another embodiment, the second node device is the second switch, then the first node device is the first switch, and the third node device is the third switch. In yet another embodiment, the second node device is a third switch, then the first node device is a second switch, and the third node device is a second computer device. The method includes but is not limited to the following steps.

S101. The first node device sends an Ethernet frame to the second node device.

S102. After the second node device receives the Ethernet frame, the second node device checks the Ethernet frame.

Exemplarily, the second node device may perform CRC check on the Ethernet frame. The FCS field of the Ethernet frame includes the CRC-32 field. The second node device may calculate the CRC value according to the payload field of the Ethernet frame according to the calculation method negotiated with the first node device. After that, compare the FCS field of the Ethernet frame with the calculated CRC value. If the calculated CRC value is different from the FCS field, the Ethernet frame verification fails; if the calculated CRC value is the same as the FCS field, the Ethernet frame verification succeeds.

By means of verification, the second node device can determine whether an error occurs during the transmission of the Ethernet frame from the first node device (ie, the previous node device) to the Ethernet link of the second node device. However, it should be noted that the second node device cannot pass the verification result to know whether an error occurred in the previous transmission process of the Ethernet frame. For example, if the Ethernet frame sent by the first node device to the second node device is originally a wrong Ethernet frame, and the wrong Ethernet frame is transmitted in the Ethernet link from the first node device to the second node device. If no error occurs during the process, then the verification of the Ethernet frame by the second node device is also successful. Exemplarily, referring to the embodiment shown in FIG. 4 , the second switch is the first node device, and the third switch is the second node device. The Ethernet frame received by the _second switch has been erroneous (containing the F2 field). Then, the original Ethernet frame sent by the _second switch to the third switch is the wrong Ethernet frame, the CRC value calculated by the third switch according to the payload field is the same as the F2 field, and the third switch successfully checks the Ethernet frame.

S103. When the Ethernet frame verification is successful, the second node device sends the Ethernet frame to the third node device.

S104. When the Ethernet frame check fails, the second node device modifies the Ethernet frame.

Specifically, the failure to check the Ethernet frame indicates that an error occurs in the Ethernet frame during the transmission process of the Ethernet link from the first node device to the second node device.

Wherein, the modified Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted. Optionally, if the first identifier is in the fourth state, it indicates that the Ethernet frame will be partially retransmitted; if the first identifier is in the fifth state, it indicates that the Ethernet frame will not be partially retransmitted.

Specifically, if the second node device determines that the Ethernet frame can be partially retransmitted, the first identifier in the modified Ethernet frame is the fourth state; if the second node device determines that the Ethernet frame cannot be partially retransmitted transmission, the first identifier in the modified Ethernet frame is the fifth state. In some embodiments, the Ethernet frame may always include the first identifier, and in the previous transmission process, the first identifier is the initial state. If the second node device fails to verify the Ethernet frame, in the case that the Ethernet frame can be partially retransmitted, the second node device modifies the first identifier from the initial state to the fourth state; when the Ethernet frame cannot be partially retransmitted In the case of transmission, the second node device modifies the first identifier from the initial state to the fifth state. In other embodiments, if the second node device fails to verify the Ethernet frame, in the case that the Ethernet frame can be partially retransmitted, the second node device adds the first identifier of the fourth state to the Ethernet frame; In the case that the Ethernet frame cannot be partially retransmitted, the second node device adds the first identifier of the fifth state to the Ethernet frame.

Exemplarily, the value of the first identifier is 1, which indicates a fourth state, indicating that the Ethernet frame will be partially retransmitted. The value of the first identifier is 0, indicating a fifth state, indicating that the Ethernet frame will not be partially retransmitted. Optionally, the value of the first identifier can also be other values, for example, 11 (represented as the fourth state, indicating that partial retransmission will be performed), 00 (represented as the fifth state, indicating that partial retransmission will not be performed) ),etc. Optionally, the first identifier may also be a preset sequence. Exemplarily, when the first identifier is the first preset sequence, it represents a fourth state, indicating that the Ethernet frame will be partially retransmitted. When the value of the first indication field is the second preset sequence, it indicates a fifth state, indicating that the Ethernet frame will not be partially retransmitted.

In addition, for the case that the first identification is always included in the Ethernet frame. In order to improve the reliability of the first identification, the first identification may include an indication byte repeated multiple times. Exemplarily, when the indication byte is 11, it indicates the fourth state, indicating that partial retransmission will be performed; when the indication byte is 00, it indicates the fifth state, indicating that partial retransmission will not be performed. The first identification includes an indication byte repeated 4 times. In this case, when the first identifier is 11111111, it represents the fourth state, indicating that the partial retransmission will be performed, and when the first identifier is 00000000, it represents the fifth state, indicating that the partial retransmission will not be performed. If an error occurs in the first identifier during the transmission process, and the first identifier is another value (for example, 11111010), it can be considered that the first identifier is in the sixth state, and the first identifier in the sixth state is used to indicate the first identifier. A logo is invalid. It can be understood that the first identifier of the fourth state and the first identifier of the fifth state may indicate that the first identifier is valid.

Further, under the premise of ensuring reliability, in order to improve the error tolerance rate of the first identifier, the first identifier can be considered valid when the correct rate of the indicated bytes is not less than a preset value (75% in an example). Wherein, the correct rate of the indication byte may be the ratio of the number of times of repeated indication bytes to the total number of times. Furthermore, whether the first flag is the fourth state or the fifth state may be determined by the value of the indication byte repeated more times. For example, the first identifier is 11111110, wherein the indication byte with a larger number of repetitions is 11, the number of repetitions is 3, and the total number of repetitions in the indication byte is 4. If the correct rate of the indicated bytes is 3/4, that is, 75%, then the first identification is deemed to be valid. If the indication byte repeated more times is 11, the first identifier is the fourth state, indicating that the Ethernet frame will be partially retransmitted.

In some embodiments, the first identifier is located at a preset position of the Ethernet frame. For example, the first identification is located after the payload field of the Ethernet frame. 6 is a schematic diagram of a frame format of an Ethernet frame provided by an embodiment of the present application. In FIG. 6, the first identifier is located before the FCS field and after the payload field (or called the data field). It should be noted that, for the transmission mode of the Ethernet frame, reference may be made to the CUT-Through mode introduced in the above content. The second node device first buffers the frame header part of the Ethernet frame (refer to FIG. 5 , the frame header part includes the destination address field, the source address field and the type/length field), and then checks the frame header part according to the destination address contained in the frame header. table, find the output port address (that is, the MAC address of the third node device) to be sent, and then send the Ethernet frame. The second node device has at least the time for buffering the frame header part to check the Ethernet frame. Since the Ethernet frame is sent in sequence according to the order of byte arrangement, the manner in which the first identifier is located after the payload field can facilitate the second node device to modify the Ethernet frame.

In addition, after the second node device fails to verify the Ethernet frame, it needs to modify the Ethernet frame, indicating that the modified Ethernet frame is invalid. Wherein, the meaning of the invalid Ethernet frame is that an error occurred in the transmission process of the Ethernet frame; the meaning of the Ethernet frame being valid is that there is no error in the transmission process of the Ethernet frame. In this way, the destination device receiving the Ethernet frame can be informed whether an error has occurred in the Ethernet frame. In some embodiments, the modified Ethernet frame further includes the second identifier, and the second identifier is used to indicate that the modified Ethernet frame is invalid.

In a possible implementation manner, after the second node device fails to verify the Ethernet frame, the second node device adds the second identifier to the Ethernet frame. That is to say, if the Ethernet frame contains the second identifier, it means that the Ethernet frame is invalid; if the Ethernet frame does not contain the second identifier, it means that the Ethernet frame is valid.

In another possible implementation manner, the Ethernet frame always includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid. Specifically, if the second identifier is in the first state, it indicates that the Ethernet frame is invalid; if the second identifier is in the second state, it indicates that the Ethernet frame is valid.

In this manner, after the second node device fails to verify the Ethernet frame, the second node device modifies the second identifier of the Ethernet frame from the second state to the first state, and the modified The second identification (ie the second identification of the first state) indicates that the modified Ethernet frame is invalid.

Exemplarily, the value of the second identifier is 1, which indicates the first state and indicates that the Ethernet frame is invalid. The value of the second identifier is 0, indicating a second state, indicating that the Ethernet frame is valid. Optionally, the value of the second identifier may also be other values, for example, 11 (represented as the first state, indicating invalid), 00 (represented as the second state, indicating valid), and so on. Optionally, the second identifier may also be a preset sequence. Exemplarily, when the second identifier is the first preset sequence, it indicates the first state, indicating that the Ethernet frame is invalid. When the second identifier is the second preset sequence, it indicates the second state, indicating that the Ethernet frame is valid.

Optionally, the second identifier may be located at a preset position of the Ethernet frame. For example, the second identification is located after the payload field (or called the data field) of the Ethernet frame. 7 is a schematic diagram of another frame format of an Ethernet frame provided by an embodiment of the present application. In FIG. 7, the second identifier is located before the FCS field and after the payload field. In addition, in this embodiment, the relative positions of the first mark and the second mark are not limited, and the first mark may be located in front of the second mark, or the first mark may also be located behind the second mark.

In other embodiments, the second node device may indicate that the Ethernet frame is invalid by inverting the FCS field of the Ethernet frame. Exemplarily, the FCS field includes a CRC value, and the CRC value calculated by the second node device is 11110000, which is different from the CRC value included in the Ethernet frame. In order to indicate that the Ethernet frame is invalid, the second node device inverts the FCS field of the Ethernet frame, and the CRC value of the FCS field of the modified Ethernet frame is 00001111.

S105. The second node device sends the modified Ethernet frame to the third node device.

In a possible implementation manner, the manner in which the second node device transmits the Ethernet frame is the CUT-Through manner. In this way, after receiving the Ethernet frame, the second node device first buffers the frame header part of the Ethernet frame, and then looks up the table according to the destination address contained in the frame header part to find the output port to be sent, and then The Ethernet frame is sent out. The manner in which the second node device checks the Ethernet frame includes: checking the Ethernet frame and sending some bits in the Ethernet frame. The manner in which the second node device sends the modified Ethernet frame includes: continuing to send bits in the modified Ethernet frame except for the partial bits. It can be understood that the partial bits of the Ethernet frame and the modified Ethernet frame are the same.

In another possible implementation manner, before the modifying the Ethernet frame, some bits of the Ethernet frame have been sent; the sending the modified Ethernet frame includes: continuing to send the Ethernet frame bits in the modified Ethernet frame except for the partial bits.

In another possible implementation manner, the sending the modified Ethernet frame includes: in the verification process, sending some bits in the modified Ethernet frame; After that, the bits in the modified Ethernet frame after the partial bits are continued to be sent.

By means of the embodiments of the present application, the first identifier may indicate whether the Ethernet frame in which the error occurs will be recovered by partial retransmission. The destination device receiving the erroneous Ethernet frame can select whether to initiate a global retransmission request according to the first identifier, which can save the transmission time of the erroneous frame and improve the retransmission efficiency of the Ethernet erroneous frame.

The following describes the manner in which the second node device determines whether the Ethernet frame can be partially retransmitted.

In a possible implementation manner, the second node device determines whether the Ethernet link between itself and the previous node device (ie, the first node device) has retransmission capability. Exemplarily, if the Ethernet link adopts an ACK/NAK (Acknowledgment/Negative Acknowledgment) error retransmission mechanism, the Ethernet link has a retransmission capability. Among them, ACK/NAK is an automatic transmission mechanism implemented by hardware, whose purpose is to ensure the effective and reliable transmission of transaction layer packets (Transaction Layer Packet, TLP). Specifically, in the process of TLP transmission, the sending device will back up each TLP to be sent in the replay buffer area (Replay Buffer). The receiving device verifies the received TLP and feeds back the receiving result to the sending device. Among them, the ACK data link layer packet (DLLP) is used to indicate that the TLP is successfully received, and the NAK DLLP is used to indicate that an error has occurred in the TLP transmission. If the sending device receives the ACK DLLP sent by the receiving device and confirms that the TLP has been successfully accepted, the backup in the Replay Buffer will be deleted. If the sending device receives the NAK DLLP sent by the receiving device and confirms that there is an error in the transmission of the TLP, the sending device will take out the data from the Replay Buffer and resend the TLP.

In another possible implementation manner, the second node device determines that it has the ability to restore the Ethernet frame. Exemplarily, the second node device may perform recovery processing on the Ethernet frame by using a forward error correction (forward error correction, FEC) algorithm or other redundant frame processing methods. The manner in which the second node device performs recovery processing on the Ethernet frame is not limited.

Referring to FIG. 8A , it is a flowchart of another method for retransmitting an Ethernet error frame provided by an embodiment of the present application. The method may be implemented based on the Ethernet system shown in FIG. 1 , and the second node device described below may be the first switch, the second switch or the third switch shown in FIG. 1 . The first node device is the previous node device of the second node device, and the third node device is the next node device of the second node device. The previous node device and the next node device are determined according to the transmission direction of the Ethernet frame. In an embodiment, the second node device is the first switch, then the first node device is the first computer device, and the third node device is the second switch. In yet another embodiment, the second node device is the second switch, then the first node device is the first switch, and the third node device is the third switch. In yet another embodiment, the second node device is a third switch, then the first node device is a second switch, and the third node device is a second computer device. The method includes but is not limited to the following steps.

S201. The first node device sends an Ethernet frame to the second node device.

For the specific implementation of this step, reference may be made to the introduction of step S101 in the above content, and details are not repeated here.

S202. After the second node device receives the Ethernet frame, the second node device checks the Ethernet frame.

For the specific implementation of this step, reference may be made to the introduction of step S102 in the above content, which will not be repeated here.

S203. When the Ethernet frame verification is successful, the second node device sends the Ethernet frame to the third node device.

S204. When the Ethernet frame check fails, the second node device modifies the Ethernet frame.

For the specific implementation of this step, reference may be made to the introduction of step S104 in the above content, which will not be repeated here.

In addition, a second identifier is always included in the Ethernet frame, and the second identifier is used to indicate whether the Ethernet frame is valid. In order to improve the reliability of the second identification, the second identification may include an indication byte repeated multiple times. Exemplarily, when the indication byte is 10, it means that the Ethernet frame is invalid; when the indication byte is 01, it means that the Ethernet frame is valid; and the second identifier includes the indication byte repeated 4 times. In this case, when the second identifier is 10101010, it represents the first state, indicating that the Ethernet frame is invalid, and when the second identifier is 01010101, it represents the second state, indicating that the Ethernet frame is valid. If an error occurs in the second identifier during transmission, and the second identifier is other values (for example, 11111010), the second identifier can be regarded as the third state, and the second identifier in the third state is used to indicate the first The second sign is invalid. It can be understood that the second identifier of the first state and the second identifier of the second state may indicate that the second identifier is valid.

Further, under the premise of ensuring reliability, in order to improve the error tolerance rate of the second identifier, the second identifier can be considered valid when the correct rate of the indicated bytes is not less than a preset value (75% in an example). Wherein, the correct rate of the indicated byte may be the ratio of the number of times of repeated indicated bytes to the total number of times. Furthermore, whether the second flag is the first state or the second state may be determined by the value of the indication byte repeated more times. For example, the second identifier is 10101011, wherein the indication byte with a larger number of repetitions is 10, the number of repetitions is 3, and the total number of repetitions in the indication byte is 4. If the correct rate of the indicated byte is 3/4, that is, 75%, then the second identification is deemed to be valid. If the indication byte repeated more times is 10, the second state is the first state, indicating that the Ethernet frame is invalid.

Optionally, the first identifier and the second identifier contained in the Ethernet frame may be combined into one indication identifier. Exemplarily, this indicator includes two indicator bytes, the first indicator byte and the first indicator have similar functions, and can be used to indicate whether the Ethernet frame will be partially retransmitted; the second indicator byte Sections and second flags have a similar effect. Can be used to indicate whether the Ethernet frame is valid. Optionally, in order to improve the reliability of the indicator, the indicator may include an indicator byte repeated multiple times.

In some embodiments, before modifying the Ethernet frame, the second node device needs to determine that the Ethernet frame is not modified. Corresponding to the manner of modifying the Ethernet frame described in the foregoing content, the second node device may determine whether the received Ethernet frame has been modified. For example, if the method of modifying the Ethernet frame is: after the node device fails to detect the Ethernet frame, modify the first identification in the Ethernet frame from the initial state to the fourth or fifth state; and modify the second identification to change from the second state to the first state. Then, the second node device can detect whether the first identifier included in the Ethernet frame is in the fourth state or the fifth state, and whether the second identifier is in the first state.

If so, it indicates that the currently received Ethernet frame has been modified (that is, the Ethernet frame is an Ethernet frame modified by the node device in the previous transmission process), then the second node device does not need to modify the Ethernet frame frame, and the operation of obtaining the retransmitted Ethernet frame, directly sending the Ethernet frame to the third node device. This is because the currently received Ethernet frame has been modified, which means that an error has occurred in the Ethernet frame in the previous transmission process, and the Ethernet frame is already an erroneous Ethernet frame at the first node device. The second node device cannot obtain the correct retransmitted Ethernet frame through its own capability or from the first node device.

If not, it indicates that the currently received Ethernet frame is not modified, then the second node device modifies the Ethernet frame and performs the operations described later. This is because the currently received Ethernet frame has not been modified, which means that no error has occurred in the Ethernet frame in the previous transmission process, and the Ethernet link between the first node device and the second node device is transmitted. The first error occurred in the process. The second node device can obtain the correct retransmitted Ethernet frame through its own capability or from the first node device. Exemplarily, for this method, reference may be made to the method flowchart shown in FIG. 8B .

It should be noted that the method of determining whether the Ethernet frame has been modified here is only an example, and corresponding to different methods of modifying the Ethernet frame, the methods of determining whether the Ethernet frame has been modified are different. In yet another example, if the way of modifying the Ethernet frame is: after the node device detects the failure of the Ethernet frame, the first identifier in the Ethernet frame is modified from the initial state to the fourth state or the fifth state; A second logo is added to the frame. Then, the second node device can detect whether the first identifier included in the Ethernet frame is in the fourth state or the fifth state, and whether the Ethernet frame includes the second identifier.

S205. The second node device sends the modified Ethernet frame to the second node device.

S206. The second node device acquires the retransmitted Ethernet frame.

It should be noted that, if the second node device determines that the Ethernet frame can be partially retransmitted, the second node device obtains the retransmitted Ethernet frame.

In a possible implementation manner, the Ethernet link between the second node device and the previous node device (ie, the first node device) has a retransmission capability. The process of acquiring the retransmitted Ethernet frame by the second node device may be as follows:

S2061. The second node device sends a partial retransmission request for the Ethernet frame to the first node device.

Exemplarily, the second node device may send a NAK packet (or referred to as NAK DLLP) to the first node device.

S2062. The second node device receives the retransmitted Ethernet frame sent by the first node device in response to the partial retransmission request.

In combination with the introduction in the above content, the first node device receives the NAK message and confirms that an error occurs in the transmission of the Ethernet frame. The first node device takes out the backup Ethernet frame from the Replay Buffer, and resends the Ethernet frame to the second node device.

In another possible implementation manner, the second node device itself has the ability to restore the Ethernet frame. The process of acquiring the retransmitted Ethernet frame by the second node device may be as follows: the second node device performs recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame. Exemplarily, the second node device may perform recovery processing on the Ethernet frame by using a forward error correction (forward error correction, FEC) algorithm or other redundant frame processing methods. The manner in which the second node device performs recovery processing on the Ethernet frame is not limited.

In some embodiments, the second node device may first perform recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame. If the retransmitted Ethernet frame fails to be checked, the second node device sends a partial retransmission request for the Ethernet frame to the first node device to obtain the retransmitted Ethernet frame sent by the first node device.

S207. The second node device checks the retransmitted Ethernet frame.

S208. When the retransmitted Ethernet frame is successfully checked, the second node device sends the retransmitted Ethernet frame to the second node device.

S210. When the retransmitted Ethernet frame fails to be checked, the second node device modifies the retransmitted Ethernet frame.

It should be noted that, for the manner in which the second node device modifies the retransmitted Ethernet frame, reference may be made to the introduction in the above content. That is to say, the modified retransmitted Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the retransmitted Ethernet frame will be partially retransmitted. The modified retransmitted Ethernet frame further includes a second identifier, and the second identifier is used to indicate that the modified retransmitted Ethernet frame is invalid.

In some embodiments, the second node device obtains a retransmitted Ethernet frame, then, in the case that the retransmitted Ethernet frame fails to check, the modified retransmitted Ethernet frame includes The first identifier is used to indicate that the retransmitted Ethernet frame will not be partially retransmitted.

In other embodiments, the second node device may try to acquire the retransmitted Ethernet frame for a preset number of times at most. Taking 3 times as an example, then, in the case that the retransmitted Ethernet frame obtained for the first and second times fails to check, the modified first identifier included in the retransmitted Ethernet frame is used. Ethernet frames that indicate retransmission will be partially retransmitted. In the case that the retransmitted Ethernet frame obtained for the third time fails to check, the modified first identifier included in the retransmitted Ethernet frame is used to indicate that the retransmitted Ethernet frame will not Retransmitted locally. In addition, it should be noted that, if the retransmitted Ethernet frame obtained in one of the multiple times is successfully verified, the second node device may not continue to obtain the retransmitted Ethernet frame.

Referring to FIG. 9A , it is a schematic diagram of another retransmission process of an Ethernet error frame provided by an embodiment of the present application. This process can be implemented based on the Ethernet architecture shown in FIG. 2 . The first computer device generates an Ethernet frame and sends the Ethernet frame to the second computer device. For the transmission process of the Ethernet, reference may be made to the above introduction, and details are not repeated here. Wherein, the initial Ethernet frame contains a second identifier and a first identifier, wherein the second identifier is 00 (second state), which is used to indicate that the Ethernet frame is valid; the first identifier is 00 (initial state).

The case of correct transmission: the first switch receives the Ethernet frame sent by the first computer device, and the first switch performs CRC check on the Ethernet frame. After the first switch successfully checks the Ethernet frame, it sends the Ethernet frame to the second switch. According to a similar process, the Ethernet frame is transmitted through the second switch and the third switch, and finally reaches the destination receiving device, that is, the second computer device. During the entire transmission process, no errors occurred in the Ethernet frame, and each node device did not modify the Ethernet frame. The second computer device may determine that the Ethernet frame is valid according to the second identifier.

When there is an error in transmission (take the Ethernet frame error code in the Ethernet link "the first computer device - the first switch" as an example): the first switch first buffers the head part of the Ethernet frame, and then according to the purpose contained in the head part Check the address table, find the output port address to be sent, and then send the Ethernet frame to the second switch. Due to a bit error, the first switch fails to verify the Ethernet frame, and the first switch determines that the Ethernet link "the first computer device - the first switch" has the retransmission capability, then the first switch modifies the first switch of the Ethernet frame. One flag is the fourth state (10, indicating that it will be partially retransmitted), and the second flag is modified to be the first state (11, indicating that it is invalid). In addition, the first switch sends a partial retransmission request (NAK message) to the first computer device to obtain the retransmitted Ethernet frame. In response to the partial retransmission request, the first computer device sends the retransmitted Ethernet frame to the first switch, and the first switch verifies that the retransmitted Ethernet frame is successful, and sends it to the next node device (ie, the second switch) The retransmitted Ethernet frame.

It should be noted that, the content shown in FIG. 9A is described by taking the example of "bit error occurs in the Ethernet frame in the Ethernet link "the first computer device - the first switch", but it is not limited to these two node devices. The implementation of the error occurred in the Ethernet link between. Exemplarily, it may also exist that the Ethernet frames are in the Ethernet links "first switch-second switch", "second switch-third switch", and "third switch-second computer device". A bit error occurs on the road. For other cases where bit errors occur, reference may be made to the description of the above example, and details are not repeated here. Assuming that no error occurs in the subsequent transmission process, the modified Ethernet frame and the retransmitted Ethernet frame will be sent to the second computer device in succession. The second computer device may determine that the Ethernet frame is invalid according to the second identifier in the modified Ethernet frame, and determine that the existing node device has performed partial retransmission according to the first identifier, so there is no need for the second computer device to report to the first computer. The device sends a global retransmission request. The second computer device discards the modified Ethernet frame without uploading to the upper layer. The second computer device waits to receive the retransmitted Ethernet frame. After receiving the retransmitted Ethernet frame, the second computer device verifies that the retransmitted Ethernet frame is successful, and determines that the retransmitted Ethernet frame is valid according to the second identifier. Next, the second computer device can upload the retransmitted Ethernet frame to the upper layer for processing.

Referring to FIG. 9B , it is a schematic diagram of another retransmission process of an Ethernet error frame provided by an embodiment of the present application. This process can be implemented based on the Ethernet architecture shown in FIG. 2 . Different from the retransmission process shown in FIG. 9A , the first switch obtains the retransmitted Ethernet frame by performing recovery processing on the Ethernet frame. For other processes, reference may be made to the introduction in the above content, which will not be repeated here.

It should be noted that, the content shown in FIG. 9B is described by taking “bit error occurs in the Ethernet frame in the Ethernet link “first computer device-first switch”” as an example, but it is not limited to these two node devices. The implementation of the error occurred in the Ethernet link between. Exemplarily, it may also exist that the Ethernet frames are in the Ethernet links "first switch-second switch", "second switch-third switch", and "third switch-second computer device". A bit error occurs on the road. For other cases where bit errors occur, reference may be made to the description of the above example, and details are not repeated here.

In addition, a case where a transmission error occurs again in the subsequent transmission process is illustrated by an example. Take the Ethernet link "second switch-third switch" as an example. The third switch first caches the head part of the Ethernet frame, then looks up the table according to the destination address contained in the head part, finds the address of the output port to be sent, and sends the Ethernet frame to the second computer device. Due to a bit error, the third switch fails to verify the Ethernet frame. The first switch determines that the Ethernet frame has been modified according to the first identifier (10) and the second identifier (11). Then, the third switch does not accept the Ethernet frame. Modification is performed, and the Ethernet frame is not retransmitted locally, and the Ethernet frame is directly sent to the second computer device.

In the following, an example is used to compare the time required to correctly transmit an Ethernet frame in the embodiment of the present application and the prior art when an error occurs in transmission. Exemplarily, the rate of each Ethernet link is 10G bits per second (bits per second, bps), the transceiver unit of each Ethernet link and the propagation delay T _link of the link are 500ns, and the CUT-Through only The 20 bytes of the Ethernet frame header need to be buffered, and the exchange delay needs to add a fixed delay of 100ns in addition to the delay of storing the Ethernet frame or the Ethernet frame header. The size of the retransmission request packet (NAK packet) is 64 bytes. The payload field carrying user messages in an Ethernet frame is 1500 bytes.

Using prior art one (refer to the introduction of the corresponding embodiment in FIG. 3 ), the total time delay T _error1 =T1+T2+T3, where T1 is the time required for the second computer device to detect the Ethernet frame packet loss, this The shortest time is the transmission time required to transmit the Ethernet frame, that is, T1=4*500+3*(1500*8/10+100)=5900ns; T2 is the required transmission time of the retransmission request packet, T2= 4*500+3*(64*8/10+100)=2453.6ns; T3 is the transmission time required to retransmit the Ethernet frame, T3=4*500+3*(1500*8/10+100) =5900ns. Then, the total time delay T _error1 =5900+2453.6+5900=14253.6ns.

If the internal blocking of transmission in a node device is considered, for example, the first switch is sending the last Ethernet frame with a length of 1500 bytes, it needs to wait at most about 1500*8/10=1200ns to send the Ethernet frame that needs to be sent currently. frame. A total of three node devices are considered for internal blocking, and the total blocking duration is T12=1200*3=3600ns. In this case, the total delay T _error1 = T1 + T2 + T3 + T12 = 5900 + 3600 + 2453.6 + 5900 = 17853.6 ns. It can be considered that the total delay of the prior art one is between 14253.6ns and 17853.6ns.

Using the second prior art (refer to the introduction of the corresponding embodiment in FIG. 4 ), the total time delay T _error2 =T4+T5+T6, where T4 is the time required for the second computer device to detect an erroneous Ethernet frame, this The transmission time required to transmit the Ethernet frame with the shortest time, namely T4=4*500+3*(20*8/10+100)=2348ns; T5 is the required transmission time of the retransmission request packet, T5=4 *500+3*(20*8/10+100)=2348ns; T6 is the transmission time required to retransmit the Ethernet frame, T6=4*500+3*(20*8/10+100)=2348ns . T _error2 =2348+2348+2348=7044ns.

Using the technology of the present invention (refer to the introduction of the corresponding embodiment in FIG. 9A ), the total time delay T _error3 = T7 + T8 , where T7 is an erroneous Ethernet frame transmitted to the first switch, and the first switch then sends a retransmission request The time required for the packet, T7=500+500=1000ns; T8 is the transmission time required to retransmit the Ethernet frame, T8=4*500+3*(20*8/10+100)=2348ns. T _error3 =1000+2348=3348ns.

It can be seen that, compared with the first prior art, the delay of the embodiment of the present application is reduced by 76.5%-81.2%; compared with the second prior art, the delay of the embodiment of the present application is reduced by 52.5%. Through the method of the embodiment of the present application, the retransmission efficiency of the Ethernet error frame can be improved.

Referring to FIG. 10 , it is a schematic diagram of another retransmission process of an Ethernet error frame provided by an embodiment of the present application. The transmission process of FIG. 10 is similar to that of FIGS. 9A and 9B, except that the Ethernet frame cannot be partially retransmitted at the first switch. Then, the first switch modifies the first flag of the Ethernet frame to the fifth state (01, indicating that it will not be partially retransmitted), and modifies the second flag to the first state (11, indicating that it is invalid).

It should be noted that the content in FIG. 10 is described by taking “bit error occurs in the Ethernet frame in the Ethernet link “the first computer device - the first switch”” as an example, but it is not limited to one of the two node devices. The implementation of the error occurred in the Ethernet link between the two. Exemplarily, it may also exist that the Ethernet frames are in the Ethernet links "first switch-second switch", "second switch-third switch", and "third switch-second computer device". A bit error occurs on the road. For other cases where bit errors occur, reference may be made to the description of the above example, and details are not repeated here.

Assuming that no error occurs in the subsequent transmission process, the modified Ethernet frame is sent to the second computer device. The second computer device may determine that the Ethernet frame is invalid according to the second identifier in the modified Ethernet frame, and determine that the Ethernet frame is not partially retransmitted according to the first identifier, then the second computer device may report to the first computer The device sends a global retransmission request. The second computer device discards the modified Ethernet frame without uploading to the upper layer. The second computer device waits to receive the retransmitted Ethernet frame. After receiving the retransmitted Ethernet frame, the second computer device verifies that the retransmitted Ethernet frame is successful, and determines that the retransmitted Ethernet frame is valid according to the second identifier in the retransmitted Ethernet frame. Next, the second computer device can upload the retransmitted Ethernet frame to the upper layer for processing.

In this case, using the technology of the present invention, the total time delay T _error4 =T9+T10+T11, where T9 is the time required for the second computer device to detect the erroneous Ethernet frame, and this time is the shortest to transmit the Ethernet frame. The transmission time required for the network frame, namely T4=4*500+3*(20*8/10+100)=2348ns; T10 is the required transmission time for the retransmission request packet, T5=4*500+3* (20*8/10+100)=2348ns; T11 is the transmission time required to retransmit the Ethernet frame, T6=4*500+3*(20*8/10+100)=2348ns. T _error4 =2348+2348+2348=7044ns.

With the first prior art, referring to the introduction in the above content, the total delay of prior art one is between 14253.6 ns and 17853.6 ns.

With the second prior art, the total time delay is comparable to that of the embodiment of the present application.

It can be seen that, compared with the first prior art, the delay of the embodiment of the present application is reduced by 50.6%-60.5%; compared with the second prior art, the delay of the embodiment of the present application is comparable. Through the method of the embodiment of the present application, the retransmission efficiency of the Ethernet error frame can be improved.

The above describes how each node device processes the Ethernet frame during the transmission of the Ethernet frame. Next, we will introduce how the destination device handles Ethernet frames. The MAC address of the destination device is the same as the MAC address contained in the destination address field in the Ethernet frame.

Referring to FIG. 11A , it is a flowchart of another method for retransmitting an Ethernet error frame provided by an embodiment of the present application. The method may be implemented based on the Ethernet system shown in FIG. 1 , and the destination device described below may be the second computer device shown in FIG. 1 . The method includes but is not limited to the following steps.

S301. A destination device receives an Ethernet frame, and checks the Ethernet frame.

S302. When the Ethernet frame verification is successful, the destination device detects whether the Ethernet frame is valid.

Wherein, whether the Ethernet frame is valid can be understood as whether an error has occurred in the Ethernet frame during the transmission process before the Ethernet link between the destination device and the previous node device. Specifically, if the Ethernet frame is valid, it means that no error has occurred in the Ethernet frame in the previous transmission process; if the Ethernet frame is invalid, it means that the Ethernet frame has encountered an error in the previous transmission process.

Corresponding to the method of modifying the Ethernet frame introduced in the above content to indicate that the Ethernet frame is invalid; the following describes the method for the destination device to detect whether the Ethernet frame is valid.

In a possible implementation manner, after the node device fails to verify the Ethernet frame, the node device adds the second identifier to the Ethernet frame. Then, the method for the destination device to detect whether the Ethernet frame is valid may be: detecting whether the Ethernet frame contains the second identifier. Specifically, if the Ethernet frame contains the second identifier, it is determined that the Ethernet frame is invalid; if the Ethernet frame does not contain the second identifier, it is determined that the Ethernet frame is valid.

The destination device can determine the length of the payload field according to the length field in the Ethernet frame. If the actually detected number of bytes of the payload field is greater than the number of bytes indicated by the length field, it can be determined that the Ethernet frame contains the The second identifier; if the actually detected number of bytes in the payload field is equal to the number of bytes indicated by the length field, it can be determined that the Ethernet frame does not contain the second identifier.

In another possible implementation manner, the Ethernet frame always includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid. The method for the destination device to detect whether the Ethernet frame is valid may be: determining whether the Ethernet frame is valid according to the second identifier. Specifically, if the second identifier is in the first state, it is determined that the Ethernet frame is invalid; if the second identifier is in the second state, it is determined that the Ethernet frame is valid.

Optionally, if the second identifier is in a third state, the second identifier in the third state is used to indicate that the second identifier is invalid. Then, it indicates that an error occurred in the previous transmission process, and the destination device determines that the Ethernet frame is invalid.

In another possible implementation manner, the node device may indicate that the Ethernet frame is invalid by inverting the FCS field of the Ethernet frame. Then, if the destination device successfully checks the Ethernet frame, it indicates that the Ethernet frame is valid.

In addition, if the verification of the Ethernet frame fails, and the reversed CRC value is the same as the CRC value included in the Ethernet frame, the destination device may determine that the Ethernet frame is invalid. If the verification of the Ethernet frame fails, and the reversed CRC value is not the same as the CRC value contained in the Ethernet frame, it indicates that the Ethernet frame is in the Ethernet link between the destination device and the previous node device. An error occurred during transfer.

S303. If the Ethernet frame is valid, upload the Ethernet frame to the upper layer for processing.

Among them, the Ethernet frame check is successful, indicating that there is no error in the transmission process of the Ethernet link "third switch-second computer device"; the Ethernet frame is valid, indicating that the Ethernet frame is in the Ethernet link. There were no errors in the respective transmissions up to the "Third Switch - Second Computer Device". Therefore, it can be determined that the Ethernet frame is correctly transmitted, and the second computer device can deliver the Ethernet frame to the upper layer for data processing.

S304. If the Ethernet frame is invalid, the destination device detects whether the Ethernet frame will be partially retransmitted.

In some embodiments, the Ethernet frame includes a first identification for indicating whether the Ethernet frame is to be partially retransmitted. The manner in which the destination device detects whether the Ethernet frame will be partially retransmitted is: determining whether the Ethernet frame will be partially retransmitted according to the first identifier. Specifically, if the first identifier is in the fourth state, it is determined that the Ethernet frame will be partially retransmitted; if the first identifier is in the fifth state, it is determined that the Ethernet frame will not be partially retransmitted. It should be noted that, for the related introduction of the first identifier, reference may be made to the introduction of the first identifier in the foregoing step S104.

In some embodiments, if the first identifier is in a sixth state, the first identifier in the sixth state is used to indicate that the first identifier is invalid. Then, it indicates that an error occurred in the first identifier in the previous transmission process, and the destination device cannot determine whether the Ethernet frame can be recovered again. Optionally, the destination device may send a global retransmission request for the Ethernet frame to the first computer device. The first computer device is a device that generates the Ethernet frame. The MAC address of the first computer device is the same as the MAC address contained in the source address field of the Ethernet frame. The global retransmission request refers to a retransmission request sent by the destination device of the Ethernet frame to the source device of the Ethernet frame.

S305. When detecting that the Ethernet frame will be partially retransmitted, the destination device waits for receiving the retransmitted Ethernet frame.

The Ethernet frame will be partially retransmitted, indicating that in the previous transmission process, the node device that received the erroneous Ethernet frame has the ability to obtain the retransmitted Ethernet frame, and the node device has tried the erroneous Ethernet frame. The network frame is partially retransmitted. In this case, the destination device does not need to rely on itself to obtain the retransmitted Ethernet frame, and can wait to receive the retransmitted Ethernet frame. For example, reference may be made to the description in the embodiment shown in FIG. 9A or FIG. 9B .

In some embodiments, if the current time is longer than the time when the Ethernet frame is received, the destination device sends a global retransmission request for the Ethernet frame to the first computer device. The preset duration may be set manually, or may be the transmission duration required to transmit one Ethernet frame calculated by the destination device. In this way, when the node device fails to retransmit the Ethernet frame locally, the destination device can initiate a global retransmission request for the Ethernet frame, so as to improve the reliability of the transmission of the Ethernet frame.

S306. When detecting that the Ethernet frame will not be partially retransmitted, the destination device sends a global retransmission request for the Ethernet frame to the first computer device.

Exemplarily, reference may be made to the description in the embodiment corresponding to FIG. 10 .

S307. When the Ethernet frame check fails, the destination device determines whether the Ethernet frame can be recovered locally.

Specifically, the failure to check the Ethernet frame indicates that an error has occurred in the Ethernet frame during the transmission process of the Ethernet link between the destination device and the previous node device. The destination device first attempts to recover the Ethernet frame.

The manner in which the destination device determines whether the Ethernet frame can be recovered locally may be: the destination device determines whether the Ethernet link between itself and the previous node device has the retransmission capability; or, the destination device determines whether it has the ability to recover the Ethernet frame. Ability.

S308. If the Ethernet frame can be recovered locally, the destination device recovers the Ethernet frame locally.

In a possible implementation manner, the Ethernet link between the destination device and the previous node device has a retransmission capability. The destination device can obtain the recovered Ethernet frame by sending a partial retransmission request to the previous node device. Afterwards, the destination device performs verification on the recovered Ethernet frame, and for subsequent operations, refer to the description of step S302 and its subsequent steps.

In another possible implementation manner, the destination device itself has the ability to restore the Ethernet frame. The second node device restores the Ethernet frame to obtain the restored Ethernet frame. After that, the destination device performs verification on the recovered Ethernet frame, and for subsequent operations, refer to the description of step S302 and its subsequent steps.

If the acquired number of failed Ethernet frame verification exceeds a preset number of times, the destination device may send a global retransmission request for the Ethernet frame to the first computer device.

S309. If the Ethernet frame cannot be recovered locally, the destination device sends a global retransmission request for the Ethernet frame to the first computer device.

In some embodiments, in the event that the Ethernet frame check fails, the destination device may also detect whether the Ethernet frame is valid before determining whether the Ethernet frame can be recovered locally. Exemplarily, for this embodiment, reference may be made to the method flowchart shown in FIG. 11B .

If it is valid, it means that the Ethernet frame has an error during the transmission process of the Ethernet link between the destination device and the previous node device, but there is no error in the previous transmission process, then the destination device determines that the Ethernet frame Whether it can be restored locally, the steps to be executed afterward can refer to the descriptions in the above-mentioned steps S308-S310.

If it is invalid, it means that the Ethernet frame has an error during the transmission process of the Ethernet link between the destination device and the previous node device, and an error also occurred during the previous transmission process, then the destination device waits to receive retransmission. Ethernet frame.

The method embodiments of the present application are described above, and the corresponding apparatus embodiments are introduced below.

Referring to FIG. 12 , it is a schematic diagram of a communication apparatus provided by an embodiment of the present application. The communication apparatus 120 may be the second node device introduced in the foregoing method embodiments. The communication device 120 may be any form of computer, server, switch, router, or network card, etc.; or a device in any form of computer, server, switch, router, or network card. The communication device 120 includes a receiving unit 1201 , a checking unit 1202 , a modifying unit 1203 and a sending unit 1204 . The receiving unit 1201 , the checking unit 1202 , the modifying unit 1203 and the sending unit 1204 will be introduced below.

The receiving unit 1201 is used for receiving Ethernet frames.

The verification unit 1202 is configured to verify the Ethernet frame.

The modifying unit 1203 is configured to modify the Ethernet frame when the Ethernet frame check fails, and the modified Ethernet frame includes a first identifier, and the first identifier is used to indicate the Ethernet frame Whether it will be retransmitted locally.

The sending unit 1204 is configured to send the modified Ethernet frame.

In some embodiments, the modified Ethernet frame further includes the second identifier, and the second identifier is used to indicate that the modified Ethernet frame is invalid.

In some embodiments, if the Ethernet frame is to be partially retransmitted, the communication device further includes an acquisition unit, the acquiring unit is configured to acquire the Ethernet frame in the case that the Ethernet frame is to be partially retransmitted the retransmitted Ethernet frame; the verification unit 1202 is further configured to verify the retransmitted Ethernet frame; the sending unit 1204 is further configured to verify the retransmitted Ethernet frame On success, the retransmitted Ethernet frame is sent.

In some embodiments, the obtaining unit is specifically configured to: send a partial retransmission request for the Ethernet frame; and receive the retransmitted Ethernet frame sent in response to the partial retransmission request.

In some embodiments, the obtaining unit is specifically configured to: perform recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame.

In some embodiments, the first identification is located after a payload field of the Ethernet frame.

In some embodiments, the checking unit is specifically configured to: check the Ethernet frame and send some bits in the Ethernet frame; the sending unit is specifically configured to: continue to send the modified bits in the Ethernet frame except the partial bits.

It should be noted that, the operations performed by each unit of the communication apparatus shown in FIG. 12 may be related to the foregoing method embodiments. It will not be described in detail here. The above-mentioned units may be implemented in hardware, software or a combination of software and hardware.

With the communication device shown in FIG. 12 , the first identifier can indicate whether the erroneous Ethernet frame will be recovered by local retransmission, and the destination device receiving the erroneous Ethernet frame can choose whether to initiate global retransmission according to the first identifier The request can save the transmission time of error frames and improve the retransmission efficiency of Ethernet error frames.

Referring to FIG. 13 , it is a schematic diagram of another communication apparatus provided by an embodiment of the present application. The communication apparatus 130 may be the destination device introduced in the foregoing method embodiments. The communication device 130 may be any form of computer, server, switch, router, or network card, or the like; or a device in any form of computer, server, switch, router, or network card. The communication device 130 includes a checking unit 1301 , a first detecting unit 1302 , a second detecting unit 1303 and a receiving unit 1304 . The verification unit 1301 , the first detection unit 1302 , the second detection unit 1303 and the receiving unit 1304 are introduced below.

The verification unit 1301 is configured to receive an Ethernet frame and perform verification on the Ethernet frame.

The first detection unit 1302 is configured to detect whether the Ethernet frame is valid when the Ethernet frame is successfully checked.

The second detection unit 1303 is configured to detect whether the Ethernet frame will be partially retransmitted if the Ethernet frame is invalid.

The receiving unit 1304 is configured to wait for receiving the retransmitted Ethernet frame when it is detected that the Ethernet frame will be partially retransmitted.

In some embodiments, the second identifier is used to indicate that the Ethernet frame is invalid, and the first detection unit 1302 is specifically configured to: detect whether the Ethernet frame contains the second identifier; if so, determine the Ethernet frame The network frame is invalid; if not included, it is determined that the Ethernet frame is valid.

In some embodiments, the Ethernet frame includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid, and the first detection unit 1302 is specifically configured to: determine according to the second identifier Whether the Ethernet frame is valid; if the second identifier is in the first state, it is determined that the Ethernet frame is invalid; if the second identifier is in the second state, it is determined that the Ethernet frame is valid.

In some embodiments, the Ethernet frame includes a first identifier, where the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted, and the second detection unit 1303 is specifically configured to: according to the The first identifier determines whether the Ethernet frame will be partially retransmitted; if the first identifier is in the fourth state, it is determined that the Ethernet frame will be partially retransmitted; if the first identifier is in the fifth state, then Make sure that Ethernet frames are not partially retransmitted.

In some embodiments, the communication device further includes a first sending unit, and the first sending unit is configured to: when it is detected that the Ethernet frame will not be partially retransmitted, send a message for the Ethernet frame Global retransmission request.

In some embodiments, the communication device further includes a determination unit and a recovery unit, the determination unit is configured to determine whether the Ethernet frame can be recovered locally when the Ethernet frame check fails; the The recovery unit is configured to perform local recovery on the Ethernet frame when it is determined that the Ethernet frame can be recovered locally.

In some embodiments, the determining unit is specifically configured to: when the Ethernet frame check fails, detect whether the Ethernet frame is valid; if the Ethernet frame is valid, determine whether the Ethernet frame is valid Can be restored locally.

In some embodiments, the receiving unit 1304 is further configured to: if the Ethernet frame is invalid, wait to receive the retransmitted Ethernet frame.

In some embodiments, the communication device further includes a second sending unit, the second sending unit is configured to: if the current time is more than a preset time period from the time when the Ethernet frame is received, send data for the Ethernet frame Global retransmission request for network frames.

It should be noted that, the operations performed by each unit of the communication apparatus shown in FIG. 13 may be related to the foregoing method embodiments. It will not be described in detail here. The above-mentioned units may be implemented in hardware, software or a combination of software and hardware.

Through the communication device shown in FIG. 13 , in the case that the Ethernet frame is detected to be invalid, it can be detected whether the Ethernet frame will be partially retransmitted. frame. It can be seen that after receiving an erroneous Ethernet frame, if it is detected that the erroneous Ethernet frame will be retransmitted locally, the destination device does not need to send a global retransmission request to the sending device of the Ethernet frame, which can save The transmission time of error frames improves the retransmission efficiency of Ethernet error frames.

Referring to FIG. 14 , it is a schematic diagram of another communication apparatus provided by an embodiment of the present application. Communication device 140 may include one or more processors 1401 . The processor 1401 may be a general-purpose processor or a special-purpose processor or the like. The processor 1401 can be used to control communication devices (computers, servers, switches, routers, network cards, etc. in any form), execute software programs, and process data of the software programs.

Optionally, the communication device 140 may include one or more memories 1402, and instructions 1404 may be stored thereon, and the instructions may be executed on the processor 1401, so that the communication device 140 executes the above method methods described in the examples. Optionally, the memory 1402 may also store data. The processor 1401 and the memory 1402 can be set independently or integrated together.

Optionally, the communication device 140 may further include a transceiver 1405 and an antenna 1406 . The transceiver 1405 may be referred to as a transceiver unit, a transceiver, or a transceiver circuit, etc., for implementing a transceiver function. The transceiver 1405 may include a receiver and a transmitter, the receiver may be called a receiver or a receiving circuit, etc., for implementing a receiving function; the transmitter may be called a transmitter or a transmitting circuit, etc., for implementing a transmitting function.

In one implementation:

The processor 1401 is configured to receive Ethernet frames through the transceiver 1405 .

The processor 1401 is further configured to verify the Ethernet frame; when the Ethernet frame verification fails, modify the Ethernet frame, and the modified Ethernet frame includes a first identifier, and the first identifier Used to indicate whether the Ethernet frame will be partially retransmitted.

The processor 1401 is further configured to transmit the modified Ethernet frame through the transceiver 1405 .

In this implementation manner, the communication apparatus 140 may be used to implement the method performed by the second node device in the foregoing method embodiment. The operations performed by the processor 1401 may be related to the foregoing method embodiments. It will not be described in detail here.

In another implementation:

The processor 1401 is further configured to verify the Ethernet frame; when the Ethernet frame verification is successful, detect whether the Ethernet frame is valid; if the Ethernet frame is invalid, detect the Ethernet frame Whether the frame will be partially retransmitted; when it is detected that the Ethernet frame will be partially retransmitted, wait to receive the retransmitted Ethernet frame.

In this implementation manner, the communication apparatus 140 can be used to implement the method performed by the destination device in the above method embodiments. The operations performed by the processor 1401 may be related to the foregoing method embodiments. It will not be described in detail here.

In another possible design, the transceiver may be a transceiver circuit, or an interface, or an interface circuit. Transceiver circuits, interfaces or interface circuits used to implement receiving and transmitting functions may be separate or integrated. The above-mentioned transceiver circuit, interface or interface circuit can be used for reading and writing code/data, or the above-mentioned transceiver circuit, interface or interface circuit can be used for signal transmission or transmission.

In another possible design, optionally, the processor 1401 may store instructions 1403, and the instructions 1403 run on the processor 1401, so that the communication apparatus 140 can execute the methods described in the above method embodiments. The instructions 1403 may be hardened in the processor 1401, in which case the processor 1401 may be implemented by hardware.

In yet another possible design, the communication apparatus 140 may include a circuit, and the circuit may implement the functions of sending or receiving or communicating in the foregoing method embodiments.

The processors and transceivers described in this application can be implemented in integrated circuits (ICs), analog ICs, radio frequency integrated circuits (RFICs), mixed-signal ICs, application specific integrated circuits (ASICs), printed circuit boards ( printed circuit board, PCB), electronic equipment, etc.

The communication device described in the above embodiments may be an access point or a station, but the scope of the communication device described in this application is not limited thereto, and the structure of the communication device may not be limited by FIG. 14 . The communication apparatus may be a stand-alone device or may be part of a larger device. For example, the communication means may be:

(1) Independent integrated circuit IC, or chip, or, chip system or subsystem;

(2) A set with one or more ICs, optionally, the IC set may also include a storage component for storing data and instructions;

(3) ASIC, such as modem (Modem);

(4) Modules that can be embedded in other equipment;

(5) Receivers, smart terminals, wireless devices, handsets, mobile units, vehicle-mounted devices, cloud devices, artificial intelligence devices, etc.;

(6) Others, etc.

For the case that the communication device may be a chip or a chip system, reference may be made to the schematic structural diagram of the chip shown in FIG. 15 . The chip 1500 shown in FIG. 15 includes a processor 1501 and an interface 1502 . The number of processors 1501 may be one or more, and the number of interfaces 1502 may be multiple.

For the case where the chip is used to implement the function of the second node device in the embodiment of the present application:

The interface 1502 is used for receiving Ethernet frames. The processor 1501 verifies the Ethernet frame; when the Ethernet frame verification fails, modifies the Ethernet frame, and the modified Ethernet frame includes a first identifier, and the first identifier Used to indicate whether the Ethernet frame will be partially retransmitted. The interface 1502 is further configured to send the modified Ethernet frame.

For the case where the chip is used to implement the function of the target device in the embodiment of the present application:

The interface 1502 is used for receiving Ethernet frames. The processor 1501 is configured to verify the Ethernet frame; when the Ethernet frame is successfully verified, detect whether the Ethernet frame is valid; if the Ethernet frame is invalid, detect the Ethernet frame Whether the Ethernet frame will be partially retransmitted; when it is detected that the Ethernet frame will be partially retransmitted, wait to receive the retransmitted Ethernet frame.

Optionally, the chip further includes a memory 1503, and the memory 1503 is used to store necessary programs and data of the terminal device.

Those skilled in the art can also understand that various illustrative logical blocks (illustrative logical blocks) and steps (steps) listed in the embodiments of the present application may be implemented by electronic hardware, computer software, or a combination of the two. Whether such functionality is implemented in hardware or software depends on the specific application and overall system design requirements. Those skilled in the art may use various methods to implement the described functions for each specific application, but such implementation should not be construed as exceeding the protection scope of the embodiments of the present application.

The present application also provides a computer-readable storage medium on which a computer program is stored, where the computer program includes program instructions, and when the program instructions are executed by a computer, the functions of any of the foregoing method embodiments are implemented.

The above-mentioned computer-readable storage medium includes, but is not limited to, flash memory, hard disk, and solid-state disk.

The present application also provides a computer program product, which implements the functions of any of the above method embodiments when the computer program product is executed by a computer.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs. When the computer program is loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer program may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line, DSL) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, a data center, or the like that includes an integration of one or more available media. The available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, high-density digital video discs (DVDs)), or semiconductor media (eg, solid state disks, SSD)) etc.

Those of ordinary skill in the art can understand that the first, second, and other numerical numbers involved in the present application are only for the convenience of description, and are not used to limit the scope of the embodiments of the present application, but also indicate the sequence.

At least one in this application may also be described as one or more, and the multiple may be two, three, four or more, which is not limited in this application. In the embodiments of the present application, for a technical feature, the technical features are distinguished by "first", "second", "third", "A", "B", "C" and "D", etc. The technical features in the "first", "second", "third", "A", "B", "C" and "D" described technical features in no order or order of magnitude.

The corresponding relationships shown in each table in this application may be configured or predefined. The values of the information in each table are only examples, and can be configured with other values, which are not limited in this application. When configuring the corresponding relationship between the information and each parameter, it is not necessarily required to configure all the corresponding relationships indicated in each table. For example, in the tables in this application, the corresponding relationships shown in some rows may not be configured. For another example, appropriate deformation adjustments can be made based on the above table, for example, splitting, merging, and so on. The names of the parameters shown in the headings in the above tables may also adopt other names that can be understood by the communication device, and the values or representations of the parameters may also be other values or representations that the communication device can understand. When the above tables are implemented, other data structures can also be used, such as arrays, queues, containers, stacks, linear lists, pointers, linked lists, trees, graphs, structures, classes, heaps, hash tables, or hash tables. Wait.

Predefined in this application may be understood as defining, predefining, storing, pre-storing, pre-negotiating, pre-configuring, curing, or pre-firing.

Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above may refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

The above are only specific implementations of the present application, but the protection scope of the present application is not limited to this. should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

A method for retransmission of Ethernet error frames, characterized in that the method comprises:

receive ethernet frames;

verifying the Ethernet frame;

When the Ethernet frame check fails, modify the Ethernet frame, and the modified Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame will be partially retransmitted;

The modified Ethernet frame is sent.
The method according to claim 1, wherein the modified Ethernet frame further includes a second identifier, and the second identifier is used to indicate that the modified Ethernet frame is invalid.
The method according to claim 1 or 2, wherein the method further comprises:

If the Ethernet frame is to be partially retransmitted, obtaining the retransmitted Ethernet frame;

Checking the retransmitted Ethernet frame;

When the retransmitted Ethernet frame is successfully checked, the retransmitted Ethernet frame is sent.
The method according to claim 3, wherein the obtaining the retransmitted Ethernet frame comprises:

sending a partial retransmission request for the Ethernet frame;

A retransmitted Ethernet frame sent in response to the partial retransmission request is received.
The method according to claim 3, wherein the obtaining the retransmitted Ethernet frame comprises:

Perform recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame.
The method according to any one of claims 1-5, wherein the first identifier is located after a payload field of the Ethernet frame.
The method according to any one of claims 1-6, wherein the checking the Ethernet frame comprises: checking the Ethernet frame and sending a part of the Ethernet frame bit;

The sending the modified Ethernet frame includes: continuing to send bits in the modified Ethernet frame after the partial bits are removed.
A method for retransmission of Ethernet error frames, characterized in that the method comprises:

receiving an Ethernet frame, and verifying the Ethernet frame;

When the Ethernet frame check is successful, detecting whether the Ethernet frame is valid;

If the Ethernet frame is invalid, detecting whether the Ethernet frame will be partially retransmitted;

When it is detected that the Ethernet frame will be partially retransmitted, it waits to receive the retransmitted Ethernet frame.
The method according to claim 8, wherein the second identifier is used to indicate that the Ethernet frame is invalid, and the detecting whether the Ethernet frame is valid comprises:

detecting whether the Ethernet frame contains a second identifier;

If included, determining that the Ethernet frame is invalid;

If not included, it is determined that the Ethernet frame is valid.
The method according to claim 8, wherein the Ethernet frame includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid, and the detecting whether the Ethernet frame is valid, include:

Determine whether the Ethernet frame is valid according to the second identifier;

If the second identifier is in the first state, determining that the Ethernet frame is invalid;

If the second identifier is in the second state, it is determined that the Ethernet frame is valid.
The method according to any one of claims 8-10, wherein the Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame is to be partially retransmitted, and the Detecting whether the Ethernet frame will be partially retransmitted, including:

determining whether the Ethernet frame will be partially retransmitted according to the first identifier;

If the first identifier is the fourth state, it is determined that the Ethernet frame will be partially retransmitted;

If the first identifier is in the fifth state, it is determined that the Ethernet frame will not be partially retransmitted.
The method according to any one of claims 8-11, wherein the method further comprises:

When it is detected that the Ethernet frame will not be locally retransmitted, a global retransmission request for the Ethernet frame is sent.
The method according to any one of claims 8-12, wherein the method further comprises:

When the Ethernet frame check fails, determining whether the Ethernet frame can be recovered locally;

When it is determined that the Ethernet frame can be recovered locally, the Ethernet frame is recovered locally.
The method of claim 13, wherein the method further comprises:

When it is determined that the Ethernet frame cannot be recovered locally, a global retransmission request for the Ethernet frame is sent.
The method according to claim 13, wherein when the Ethernet frame check fails, determining whether the Ethernet frame can be recovered locally comprises:

When the Ethernet frame check fails, detecting whether the Ethernet frame is valid;

If the Ethernet frame is valid, it is determined whether the Ethernet frame can be recovered locally.
The method of claim 15, wherein the method further comprises:

If the Ethernet frame is invalid, waiting to receive the retransmitted Ethernet frame.
The method according to any one of claims 8-16, characterized in that after the waiting to receive the retransmitted Ethernet frame, the method further comprises:

If the current time is more than a preset time period from the time when the Ethernet frame is received, a global retransmission request for the Ethernet frame is sent.
A communication device, characterized in that the communication device comprises a receiving unit, a checking unit, a modifying unit and a sending unit, wherein:

the receiving unit for receiving Ethernet frames;

the verification unit, configured to verify the Ethernet frame;

The modification unit is configured to modify the Ethernet frame when the Ethernet frame check fails, and the modified Ethernet frame includes a first identification, and the first identification is used to indicate whether the Ethernet frame is will be partially retransmitted;

The sending unit is configured to send the modified Ethernet frame.
The communication device according to claim 18, wherein the modified Ethernet frame further comprises a second identification, and the second identification is used to indicate that the modified Ethernet frame is invalid.
The communication device according to claim 18 or 19, characterized in that, the communication device further comprises an acquisition unit,

The obtaining unit is configured to obtain the retransmitted Ethernet frame when the Ethernet frame will be partially retransmitted;

The verification unit is further configured to verify the retransmitted Ethernet frame;

The sending unit is further configured to send the retransmitted Ethernet frame when the retransmitted Ethernet frame is successfully checked.
The communication device according to claim 20, wherein the obtaining unit is specifically configured to:

sending a partial retransmission request for the Ethernet frame;

A retransmitted Ethernet frame sent in response to the partial retransmission request is received.
The communication device according to claim 20, wherein the obtaining unit is specifically configured to:

Perform recovery processing on the Ethernet frame to obtain the retransmitted Ethernet frame.
The communication device according to any one of claims 18-22, wherein the first identifier is located after a payload field of the Ethernet frame.
The communication device according to any one of claims 18-23, wherein the check unit is specifically configured to: check the Ethernet frame and send some bits in the Ethernet frame;

The sending unit is specifically configured to: continue to send bits in the modified Ethernet frame after the partial bits are removed.
A communication device, characterized in that the communication device comprises a verification unit, a first detection unit, a second detection unit and a receiving unit, wherein:

The verification unit is configured to receive an Ethernet frame and verify the Ethernet frame;

the first detection unit, configured to detect whether the Ethernet frame is valid when the Ethernet frame verification is successful;

the second detection unit, configured to detect whether the Ethernet frame will be partially retransmitted if the Ethernet frame is invalid;

The receiving unit is configured to wait to receive the retransmitted Ethernet frame when it is detected that the Ethernet frame will be partially retransmitted.
The communication device according to claim 25, wherein the second identifier is used to indicate that the Ethernet frame is invalid, and the first detection unit is specifically configured to:

detecting whether the Ethernet frame contains a second identifier;

If included, determining that the Ethernet frame is invalid;

If not included, it is determined that the Ethernet frame is valid.
The communication device according to claim 25, wherein the Ethernet frame includes a second identifier, and the second identifier is used to indicate whether the Ethernet frame is valid, and the first detection unit is specifically configured to:

Determine whether the Ethernet frame is valid according to the second identifier;

If the second identifier is in the first state, determining that the Ethernet frame is invalid;

If the second identifier is in the second state, it is determined that the Ethernet frame is valid.
The communication device according to any one of claims 25-27, wherein the Ethernet frame includes a first identifier, and the first identifier is used to indicate whether the Ethernet frame is to be partially retransmitted, so The second detection unit is specifically used for:

determining whether the Ethernet frame will be partially retransmitted according to the first identifier;

If the first identifier is the fourth state, it is determined that the Ethernet frame will be partially retransmitted;

If the first identifier is in the fifth state, it is determined that the Ethernet frame will not be partially retransmitted.
The communication device according to any one of claims 25-28, wherein the communication device further comprises a first sending unit, and the first sending unit is configured to:

When it is detected that the Ethernet frame will not be locally retransmitted, a global retransmission request for the Ethernet frame is sent.
The communication device according to any one of claims 25-29, wherein the communication device further comprises a determination unit and a recovery unit,

The determining unit is configured to determine whether the Ethernet frame can be recovered locally when the Ethernet frame check fails;

The recovery unit is configured to perform local recovery on the Ethernet frame when it is determined that the Ethernet frame can be recovered locally.
The communication device according to claim 30, wherein the determining unit is specifically configured to:

When the Ethernet frame check fails, detecting whether the Ethernet frame is valid;

If the Ethernet frame is valid, it is determined whether the Ethernet frame can be recovered locally.
The communication device according to claim 31, wherein the receiving unit is further configured to:

If the Ethernet frame is invalid, waiting to receive the retransmitted Ethernet frame.
The communication device according to any one of claims 25-32, wherein the communication device further comprises a second sending unit, and the second sending unit is configured to:

If the current time is more than a preset time period from the time when the Ethernet frame is received, a global retransmission request for the Ethernet frame is sent.
A communication device, comprising a processor, a memory and a transceiver;

the transceiver for receiving Ethernet frames;

the memory for storing program codes;

The processor is configured to call the program code from the memory to execute the method according to any one of claims 1-7.
A communication device, comprising a processor, a memory and a transceiver;

the transceiver for receiving Ethernet frames;

the memory for storing program codes;

The processor is configured to call the program code from the memory to execute the method according to any one of claims 8-17.
A computer-readable storage medium, characterized in that, the computer-readable storage medium is used for storing instructions, and when the instructions are executed, the method according to any one of claims 1-7 is implemented.
A computer-readable storage medium, characterized in that, the computer-readable storage medium is used for storing instructions, which, when executed, enable the method according to any one of claims 8 to 17 to be implemented.
A communication device for performing the method of any one of claims 1-7.
A communication device for performing the method of any one of claims 8-17.
A computer program product comprising instructions which, when run on a computer, cause the method of any of claims 1-7 to be implemented; or the method of any of claims 8-17 to be implemented .
A computer program which, when run on a computer, causes the method of any one of claims 1-7 to be implemented; or the method of any one of claims 8-17 to be implemented.
A communication system, comprising the communication device according to any one of claims 18-24 and the communication device according to any one of claims 25-33.