CN111817818A - Method and device for transmitting data packet - Google Patents

Abstract

The invention discloses a method and a device for transmitting data packets, which are used for solving the problem of high packet loss rate of the transmitted data packets. The scheme provided by the application comprises the following steps: determining at least one target block in a user layer according to at least one data packet received by a network layer and an annular preset block queue in a kernel layer, wherein when the number of the determined target blocks is multiple, the multiple target blocks are adjacent in the preset block queue; storing the at least one data packet into the at least one target block. The scheme of the embodiment of the invention can ensure that the target block for storing the data packet conforms to the preset block queue, so that the physical addresses of the stored data packet are logically continuous, the problem of packet loss caused by logic address dispersion is avoided, and the probability of packet loss is reduced.

Description

A method and apparatus for transmitting data packets

technical field

The present invention relates to the field of data transmission, in particular to a method and device for transmitting data packets.

Background technique

In the Linux operating system, I/O operations enable data to be transferred between a buffer in the operating system's kernel address space and a buffer defined in the application's address space. In the process of data transmission, the CPU overhead is large, and the data transmission capacity is low.

There is a method for zero-copy data transmission, which may include the following two steps:

1) Create an effective user-level communication interface, that is, the application program directly sends or receives data from the communication interface, eliminating unnecessary copying processes in the system kernel;

2) Direct data transmission from the input end to the output end of the router, that is, the received message is only buffered once in the memory, and the message in the buffer queue is directly sent to the output port after processing the necessary control information to realize the message Fast forwarding.

This existing zero-copy data transmission method usually randomly stores data packets, so that the locations where data packets are stored are scattered, and packet loss is prone to occur during transmission.

How to reduce the packet loss rate of transmission data packets is the technical problem to be solved by this application.

SUMMARY OF THE INVENTION

The purpose of the embodiments of the present application is to provide a method and apparatus for transmitting data packets, so as to solve the problem of high packet loss rate of transmitting data packets.

In a first aspect, a method for transmitting data packets is provided, including:

At least one target block in the user layer is determined according to at least one data packet received by the network layer and a ring-shaped preset block queue in the kernel layer, wherein when the number of the determined target blocks is multiple, multiple the target block is adjacent in the preset block queue;

The at least one data packet is stored in the at least one target block.

In a second aspect, an apparatus for transmitting data packets, comprising:

The determining module determines at least one target block in the user layer according to at least one data packet received by the network layer and the annular preset block queue in the kernel layer, wherein, when the number of the determined target blocks is multiple , a plurality of the target blocks are adjacent in the preset block queue;

The storage module stores the at least one data packet in the at least one target block.

In a third aspect, an electronic device is provided, comprising a processor, a memory, and a computer program stored on the memory and executable on the processor, the computer program being executed by the processor to implement the method described in the first aspect steps of the method.

In a fourth aspect, a computer-readable storage medium is provided, and a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method of the first aspect are implemented.

In the embodiment of the present application, at least one target block in the user layer is determined according to at least one data packet received by the network layer and a ring-shaped preset block queue in the kernel layer, wherein when the determined number of target blocks is When there are more than one, a plurality of the target blocks are adjacent in the preset block queue; and the at least one data packet is stored in the at least one target block. The solution of the embodiment of the present invention can ensure that the target block of the stored data packet conforms to the preset block queue, so that the physical addresses of the stored data packet are logically continuous, avoid the problem of packet loss caused by the dispersion of logical addresses, and reduce the probability of packet loss. .

Description of drawings

The accompanying drawings described herein are used to provide further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the attached image:

Fig. 1a is one of the schematic flow charts of a method for transmitting data packets of the present application;

1b is a schematic diagram of the logical structure of a transmission data packet of the present application;

1c is a schematic diagram of the logical structure of the preset block queue of the present application;

2 is the second schematic flow diagram of a method for transmitting a data packet of the present application;

3 is the third schematic flow chart of a method for transmitting data packets according to the present application;

4 is the fourth schematic flow chart of a method for transmitting data packets according to the present application;

5 is the fifth schematic flow chart of a method for transmitting data packets according to the present application;

6 is the sixth schematic flow chart of a method for transmitting data packets of the present application;

7 is the seventh schematic flow chart of a method for transmitting data packets according to the present application;

FIG. 8 is the eighth schematic flow chart of a method for transmitting data packets according to the present application;

FIG. 9 is a schematic structural diagram of an electronic device of the present application.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. The numbers of the drawings in this application are only used to distinguish each step in the solution, and are not used to limit the execution sequence of each step, and the specific execution sequence is subject to the description in the specification.

In the prior art, a typical zero-copy data transmission architecture includes a network card layer, a kernel layer, and a user layer. In the zero-copy scheme, the user layer interacts directly with the network card layer, which avoids the resource overhead required for replication and transmission at the kernel layer and achieves the goal of zero-copy.

Due to the large number of data packets and large capacity in the current high-speed network environment, the existing zero-copy technology adopts a random storage method for the captured data packets, so the storage locations of data packets are often scattered, which is very important for intrusion detection. Subsequent work reading packets and analyzing packets creates inconvenience. In addition, since the data packets are stored in a scattered manner, the probability of packet loss is often high, and the data packets finally obtained by the user process may be incomplete.

In view of the problems existing in the prior art, the present application provides a method for transmitting data packets, as shown in Figure 1a, comprising the following steps:

S11: Determine at least one target block in the user layer according to the at least one data packet received by the network layer and the annular preset block queue in the kernel layer, wherein, when the number of the determined target blocks is multiple, A plurality of the target blocks are adjacent in the preset block queue;

S12: Store the at least one data packet in the at least one target block.

Referring to the structure shown in FIG. 1b, the solution provided in this application is suitable for a transmission architecture including a network layer Z, a kernel layer Y and a user layer X.

The user layer X (User) includes at least one buffer N. When the user layer X includes multiple buffers N, the size of each buffer N may be different, and the physical locations of the multiple buffers N may be discontinuous. According to the positions and sizes of the multiple buffers N, the buffer N in the user layer X may be divided into multiple buffer blocks in advance. Preferably, the capacity of each buffer block is equal. Subsequently, a unique descriptor can be established for each buffer block, which is used to characterize the buffer block.

In addition, the user layer X in this solution can also be regarded as a part of the application program, and the buffer N in the user layer X can be an area associated with the application program that can store data.

The function of the kernel layer Y (Kernel) is to realize the data transmission between the network layer and the user layer X. A ring-shaped preset block queue M is preset in the kernel layer Y, and the preset block queue M may also be called an idle queue. The preset block queue M includes a plurality of identifiers that can represent buffer blocks in the user layer X that can store data. Specifically, these identifiers may be the descriptors of the above-mentioned blocks. Each item in the above-mentioned predetermined block queue M includes a descriptor of the above-mentioned block. Preferably, there will not be two or more identical block descriptors in the preset block queue M.

The network layer Z (Network Information Center, NIC) may contain a receive ring (not shown in FIG. 1b ), and the content of the receive ring is synchronized with the preset block queue M in the kernel layer Y. The network layer Z directly transmits the data packet O received from the network to the buffer of the user layer X according to the information in the receiving ring.

Specifically, the number of target blocks can be determined according to the capacity of the data packet O received in the network layer Z, and when the capacity of the data packet is large, the determined number of target blocks can be larger. After the number of target blocks is determined, descriptors of the above-mentioned number of target blocks can be selected from the preset block queue M. When the above-mentioned number is greater than 1, the descriptors selected from the preset block queue M are adjacent.

As shown in FIG. 1c, the figure shows a schematic diagram of the logical structure of the preset block queue. The preset block queue includes m ₁ , m ₂ , m ₃ , m ₄ , m ₅ , m ₆ , m ₇ , and m ₈ , a total of 8 descriptors corresponding to the buffer blocks.

Assuming that the network layer determines that the number of target blocks is 1 according to the received data packet, the selected descriptor may be m ₁ based on the preset block queue shown in FIG. 1c . Assuming that the network layer determines that the number of target blocks is 3 according to the received data packets, then based on the preset block queue shown in FIG. 1c , the selected descriptors may be m ₁ , m ₂ , and m ₃ . In fact, the starting position of the selection descriptor can be preset in the preset block queue according to the actual situation. For example, the position of the m ₁ descriptor shown in the dashed box shown in FIG. 1c is set as the starting position of the selection descriptor, then according to the determined number of target blocks, from the position shown in the dashed box Picks consecutive descriptors along a preset direction. In this embodiment, the preset direction may be a clockwise direction.

In the actual application process, the network layer can continuously receive multiple data packets. It is assumed that the network layer receives three packets of O ₁ , O ₂ , and O ₃ in sequence. Meanwhile, assuming that each block in the preset block queue can store one data packet, the number of target blocks determined is three. Based on the preset block queue structure shown in FIG. 1 c , the descriptors of the three selected target blocks may be m ₁ , m ₂ , and m ₃ . Subsequently, the network layer can transmit the data packet _O1 to the target block corresponding to the descriptor _m1 , transmit the data packet _O2 to the target block corresponding to the descriptor _m2 , and transmit the data packet _O3 to the descriptor within the target block corresponding to _m3 .

In addition, the network layer may also receive the data packet O _{1 first} , transmit the data packet O ₁ to the target block corresponding to the descriptor m ₁ , and then receive the data packet O ₂ .

In the solution provided by the present application, after at least one target block is determined, the network layer directly transmits the data packet to the above-mentioned at least one target block in the user layer. It can ensure that the target block of the stored data packet conforms to the preset block queue, so that the physical address of the stored data packet is logically continuous, and the application can determine the physical location of the stored data packet based on the preset block queue to avoid logical address dispersion The resulting packet loss problem reduces the probability of packet loss. In contrast, in the prior art, the network layer first transmits the data packet from the storage unit of the network layer to a certain buffer of the kernel, and then the interrupt handler pushes the data packet in the kernel layer into the upper-layer protocol stack. The disadvantage of this way of transferring data mainly lies in multiple copies and system calls. In the solution provided by this application, the network layer transmits data packets according to the receiving ring, and can determine the target block located at the user layer according to the descriptor in the preset block queue, and then directly transmits the data packets from the network layer to the user layer. No longer enter the kernel's protocol stack.

Preferably, before performing the above steps, it is necessary to perform initialization operations on the user layer, the kernel layer and the network layer. Based on the method provided by the foregoing embodiment, before performing the foregoing step S11, as shown in FIG. 2, the method provided by the present application further includes:

S101: establish at least one block at the user layer, and generate a block address mapping table that includes the mapping relationship between the at least one block and the physical address of the at least one block;

S102: Generate the preset block queue in the kernel layer according to the established at least one block.

Specifically, an initialization operation is performed on the user layer first. First, at least one buffer in the user layer that can be used to store data packets is determined, and the at least one buffer is divided into multiple buffer blocks (UserBuffer). The size and number of buffer blocks can be preset according to actual needs. Subsequently, the address mapping table of the user layer is initialized, and the size of the address mapping table is related to the number of buffer blocks. The address mapping table includes the above-mentioned correspondence between the buffer blocks and the physical addresses of the buffer blocks.

After initializing the user layer, perform initialization operations on the kernel layer. A ring-shaped preset block queue (FREEQ) is established, and each item in the queue corresponds to a buffer block in the user layer that can store data. Specifically, define FREEQ as a free linked list, initialize FREEQ, and assign FREE[i]=i; make the pointer of FREE(i) point to UserBuffer[i].

The above solution provided by the present application can realize the initialization operation on the user layer and the kernel layer, so that the preset block queue in the kernel layer is associated with the blocks in the user layer. According to the preset block queue in the kernel layer, it can be known which blocks in the user layer can be used to store data, so that the network layer can store the data packets in the free buffer blocks in the user layer.

Based on the method provided by the foregoing embodiment, preferably, as shown in FIG. 3 , the foregoing step S102 includes:

S1021: Generate a descriptor corresponding to the at least one block one-to-one, and the descriptor includes a block symbol of the block corresponding to the descriptor;

S1022: Generate a preset block queue including at least one descriptor in the kernel layer.

Specifically, each buffer block in the user layer corresponds to a block symbol, and each block symbol is filled into the offset field of the descriptor corresponding to the buffer block. Subsequently, these descriptors corresponding to the buffer blocks are arranged in a circular arrangement into the above-mentioned preset block queue (FREEQ), so that each item in the preset block queue corresponds to a user layer that can be used to store data. buffer block.

The above solution provided by the present application can generate a descriptor corresponding to the buffer block based on the block identifier of the buffer block, so that each item in the preset block queue corresponds to a buffer block in the user layer.

Based on the method provided by the above embodiment, preferably, after the above step S102, as shown in FIG. 4, the method provided by the present application further includes:

S103: generate a physical address table in the kernel layer according to the established at least one block, and the physical address table includes a correspondence between the physical addresses of each established block and each established block;

S104: Generate a data field in the network layer according to the preset block queue and the physical address table, where the data field includes a pointer to the preset block queue and an address of the physical address table.

Specifically, in the process of initializing the kernel layer, a physical address table may also be established in the kernel layer, and the physical address table is associated with the above-mentioned preset block queue. The physical address table includes the correspondence between the buffer blocks and the physical addresses of the buffer blocks.

The above solution provided by the present application can establish a physical address table associated with the preset block queue in the kernel layer, and the physical address table includes the physical address of each block in the preset block queue, so that the network layer can use the preset block queue according to the preset block queue. Let the block queue and the physical address table determine the buffer block that can be used to store data, and determine the physical address of the buffer block, and then transmit the received data.

Then, initialize the network layer. The data field is initialized in the network layer, and the pointer of the preset block queue in the kernel layer and the address in the physical address table are filled into the data field in the network layer. The network layer can determine the buffer block and the physical address of the block that can be used to store data according to the preset block queue and the physical address table in the kernel layer.

Based on the method provided by the foregoing embodiment, preferably, before the foregoing step S12, as shown in FIG. 5, the method further includes:

S105: Write the size of at least one data packet received by the network layer into the data field;

Wherein, the above-mentioned step S12 includes:

S121: Determine the descriptor of the target block in the preset block queue according to the data field of the network layer;

S122: Determine the physical address corresponding to the descriptor of the target block in the physical address table according to the data field of the network layer;

S123: Store the at least one data packet in the target block according to the size of the at least one data packet in the data field of the network layer.

Specifically, after initializing the user layer, the kernel layer and the network layer, the network layer performs packet capture. After the data packet to be stored reaches the network layer, a hardware interrupt is issued, and the packet receiving interrupt handler.

Then, based on the data field of the network layer, the descriptor of the available target block in the preset block queue is determined, and the descriptor is parsed to obtain the block number and physical address of the target block. Next, the physical address of the target block is written into the data field of the network layer, and the data packet is transmitted based on the size of the data packet.

The solution provided by this application can directly transmit the data packet received by the network layer to the buffer block in the user layer. During the transmission process, the network layer determines at least one target block that can be used to store the received data packets according to the preset block queue and physical address list in the kernel layer, to ensure that the determined target block can be used to store the received data packets. to avoid packet loss or overflow during transmission.

After the network layer transmits the data packet to the target block in the user layer, the application program in the user layer can parse the data packet stored in the target block. Specifically, the application program of the user layer can monitor the number of descriptors in the preset block queue. Once the number of descriptors in the preset block queue changes, it indicates that there are buffer blocks in the user layer that store new data. The application program of the user layer can find the target block address where the data is stored through the pointer in the block address mapping table, and then parse the data in the target block of the address.

The technical solution provided by the present application can directly transmit data packets from the network layer to the user layer, so that the application program in the user layer can directly parse the data in the target block. In the process of transmission, data packets do not need to enter the kernel layer to achieve zero copy. Moreover, the solution provided by the present application can effectively reduce the burden of the system, improve the efficiency of packet capture, and avoid situations such as packet loss and overflow. . Based on the method provided by the above embodiment, preferably, the preset block queue includes a descriptor of at least one idle block;

As shown in Figure 6, the above step S11 includes:

S111: Determine the size of the data packet to be stored according to at least one data packet received by the network layer;

S112: determine the number of target blocks according to the size of the data packet to be stored;

S113: Determine the physical address of each target block in the user layer according to the descriptor of the target block in the preset block queue.

In the embodiments provided in this application, the network layer may receive multiple data packets in advance, and then transmit the multiple data packets to the target block in the user layer. The size of the data packet to be stored may be the sum of the capacity of at least one data packet received by the network layer.

Preferably, the buffer in the user layer is divided into several buffer blocks of equal size in advance, and the number of target blocks is determined according to the size of the divided buffer block and the size of the data packet to be stored. Make the total capacity of the determined number of target blocks greater than or equal to the size of the data packet to be stored.

The above-mentioned number of block descriptors are taken out from the preset block queue, the physical addresses of the above-mentioned target blocks are determined according to these descriptors, and the network layer transmits at least one received data packet to the In the target block of the user layer.

Through the technical solution provided in this application, the network layer can realize batch transmission of multiple data packets, and the determined number of target blocks is adapted to the data packets to be stored, so as to avoid data overflow or the number of target blocks caused by too few target blocks. Excessive waste of resources.

Based on the method provided by the foregoing embodiment, preferably, the foregoing step S12 includes:

When the number of determined target blocks is multiple, according to the physical address of each target block in the user layer, the data packets to be stored are stored in the order of the preset block queue to in multiple target blocks determined.

In the embodiment of the present application, the determined number of target blocks is multiple, and the network layer transmits the data packets to the target blocks of the user layer according to the sequence of the preset block queue. Specifically, it is assumed that the network layer receives two data packets, the size of the first data packet is 17.3M, and the size of the second data packet is 8.2M. The preset capacity of each cache block is 10M, then according to the size of the block to be stored, the number of target blocks is determined to be 3, which are the first target block, the second target block and the third target block. . Descriptors of three target blocks are obtained from the preset block queue, and physical addresses of the three target blocks are determined according to the obtained descriptors. Then, the first data packet is divided into two parts, the first part is 10M, and the second part is 7.3M. The first part of the first data packet is transferred into the first target block, the second part of the first data packet is transferred into the second target block, and the second data packet is transferred into the third target block, thereby Realize the transmission of data packets.

Through the solution provided by the present application, the data packets received by the network layer can be transmitted and stored in the target block of the user layer according to the sequence of the preset block queue. It is convenient for the application program to determine the order of reading target blocks according to the preset block queue, to avoid the application program from reading multiple incomplete data packets, and to improve the efficiency of the application program to read the data packets.

Based on the method provided by the above-mentioned embodiment, preferably, after the above-mentioned step S12, as shown in FIG. 7, the method further includes:

S114: write the physical address of the at least one target block storing the at least one data packet into the block address mapping table of the user layer;

S115: The application parses the at least one data packet according to the block address mapping table of the user layer.

In this embodiment, the user layer further includes a block address mapping table, and the block address mapping table includes a corresponding relationship between the target block storing the data packet and the physical address of the target block. The application can determine the target block in which the data packet has been stored and its physical address according to the block address mapping table, so as to read the physical address of the target block and perform the analysis of the data packet.

Based on the method provided by the above embodiment, preferably, after the above step S12, as shown in FIG. 8, the method provided by the present application further includes:

S116: Remove the descriptor of the at least one target block storing the at least one data packet in the preset block queue;

After the above step S115, it also includes:

S117: Add a descriptor for storing at least one target block of the at least one data packet to the preset block queue.

In this embodiment, the descriptor of the target block in which the data packet is stored is removed from the preset block queue, so that the buffer blocks represented by the descriptors in the preset block queue are all buffer blocks that do not store data. After the application parses the data packets in the target block, it adds the descriptor of the target block that originally stored the data packet back to the preset block queue, so that the target block queue can continue to store the new data received by the network layer. data pack.

The solution provided by this application can remove the descriptor of the target block from the preset block queue after the target block stores a new data packet, so as to avoid the data packet stored in the target block from being parsed by a new application program. Data coverage. In addition, after the application program parses the data in the target block, the descriptor of the target block is added back to the preset block queue to ensure that the preset block queue always has the descriptor of the target block capable of storing data.

In order to solve the above problems existing in the prior art, the present application also provides a device 90 for transmitting data packets, as shown in FIG. 9 , including:

The determining module 91 determines at least one target block in the user layer according to at least one data packet received by the network layer and the annular preset block queue in the kernel layer, wherein when the number of the determined target blocks is multiple , a plurality of the target blocks are adjacent in the preset block queue;

The storage module 92 stores the at least one data packet in the at least one target block.

In the embodiment of the present application, at least one target block in the user layer is determined according to at least one data packet received by the network layer and a ring-shaped preset block queue in the kernel layer, wherein when the determined number of target blocks is When there are more than one, a plurality of the target blocks are adjacent in the preset block queue; and the at least one data packet is stored in the at least one target block. The solution of the embodiment of the present invention can ensure that the target block of the stored data packet conforms to the preset block queue, so that the physical addresses of the stored data packet are logically continuous, avoid the problem of packet loss caused by the dispersion of logical addresses, and reduce the probability of packet loss. .

Preferably, an embodiment of the present invention further provides an electronic device, including a processor, a memory, a computer program stored in the memory and running on the processor, and the computer program implements the above-mentioned image when executed by the processor The various processes of the processing method embodiments can achieve the same technical effect, and are not repeated here to avoid repetition.

Embodiments of the present invention further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, each process of the above-mentioned image processing method embodiment can be achieved, and can achieve the same In order to avoid repetition, the technical effect will not be repeated here. The computer-readable storage medium is, for example, a read-only memory (Read-Only Memory, ROM for short), a random access memory (Random Access Memory, RAM for short), a magnetic disk, or an optical disk.

It should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements, It also includes other elements not expressly listed or inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

From the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform, and of course hardware can also be used, but in many cases the former is better implementation. Based on this understanding, the technical solutions of the present invention can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products are stored in a storage medium (such as ROM/RAM, magnetic disk, CD-ROM), including several instructions to make a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present invention.

The embodiments of the present invention have been described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific embodiments, which are merely illustrative rather than restrictive. Under the inspiration of the present invention, without departing from the spirit of the present invention and the scope protected by the claims, many forms can be made, which all belong to the protection of the present invention.

Claims (10)

1. a method for transmitting data packet, is characterized in that, comprises: At least one target block in the user layer is determined according to at least one data packet received by the network layer and a ring-shaped preset block queue in the kernel layer, wherein when the number of the determined target blocks is multiple, multiple the target block is adjacent in the preset block queue; The at least one data packet is stored in the at least one target block. 2. The method of claim 1, wherein the preset block queue comprises a descriptor of at least one idle block; The said at least one target block in the user layer is determined according to the at least one data packet received by the network layer and the annular preset block queue in the kernel layer, including: Determine the size of the data packet to be stored according to at least one data packet received by the network layer; Determine the number of target blocks according to the size of the data packet to be stored; The physical address of each target block in the user layer is determined according to the descriptor of the target block in the preset block queue. 3. The method of claim 2, wherein the storing the at least one data packet in the at least one target block comprises: When the number of determined target blocks is multiple, according to the physical address of each target block in the user layer, the data packets to be stored are stored in the order of the preset block queue to in multiple target blocks determined. 4. The method of claim 1, further comprising: after storing the at least one data packet into the at least one target block: Writing the physical address of the at least one target block storing the at least one data packet into the block address mapping table of the user layer; The application parses the at least one data packet according to the block address mapping table of the user layer. 5. The method of claim 4, further comprising: after storing the at least one data packet into the at least one target block: removing the descriptor of the at least one target block storing the at least one data packet in the preset block queue; After the application parses the at least one data packet according to the block address mapping table of the user layer, it further includes: A descriptor for storing at least one target block of the at least one data packet is added to the preset block queue. 6. method as claimed in claim 1 is characterized in that, before at least one target block in user layer is determined according to at least one data packet received by network layer and the annular preset block queue in kernel layer, Also includes: Establish at least one block at the user layer, and generate a block address mapping table that includes the mapping relationship between the at least one block and the physical address of the at least one block; The preset block queue is generated in the kernel layer according to the established at least one block. 7. The method of claim 6, wherein generating the preset block queue in the kernel layer according to the at least one block established, comprising: generating a descriptor corresponding to the at least one block one-to-one, the descriptor comprising a block symbol of the block corresponding to the descriptor; A preset block queue containing at least one descriptor is generated in the kernel layer. 8. The method of claim 7, wherein after generating the preset block queue in the kernel layer according to the at least one block established, further comprising: According to the established at least one block, a physical address table is generated in the kernel layer, and the physical address table includes a correspondence between each established block and the physical address of each established block; A data field is generated in the network layer according to the preset block queue and the physical address table, and the data field contains a pointer to the preset block queue and an address of the physical address table. 9. The method of claim 8, further comprising: before storing the at least one data packet into the at least one target block: writing the size of at least one data packet received by the network layer into the data field; Wherein, the storing the at least one data packet in the at least one target block includes: According to the data field of the network layer, determine the descriptor of the target block in the preset block queue; According to the data field of the network layer, determine the physical address corresponding to the descriptor of the target block in the physical address table; The at least one data packet is stored in the target block according to the size of the at least one data packet in the data field of the network layer. 10. A device for transmitting data packets, comprising: The determining module determines at least one target block in the user layer according to at least one data packet received by the network layer and a ring-shaped preset block queue in the kernel layer, wherein, when the number of the determined target blocks is multiple , a plurality of the target blocks are adjacent in the preset block queue; The storage module stores the at least one data packet in the at least one target block.

Images