Data Packet Association Data Processing Method and System

Abstract

The present application provides a method and system for processing packet associated data. The method includes: classifying the associated data processing instructions for the target packet generated by itself to obtain target associated data processing instructions; sending the target associated data processing instructions to the corresponding data processing engine, so that the data processing engine converts the target associated data processing instructions into the target address of the target program segment included in the data processing engine, and executes the target program segment to read, update or write back the target associated data in the storage management engine. The present application can effectively ensure the atomicity of packet associated data processing, and at the same time can effectively improve the concurrent working efficiency of the data processing engine.

Description

Technical Field

[0001] This application relates to the field of network data processing, and particularly to a method and system for processing associated data of data packets. Background Art

[0002] In the processing flow of a network processor, network data packets enter the processor for processing and leave the network processor after the processing is completed. During the processing, general processes such as packet parsing, table lookup, and action execution are included. Most of the processing processes have no association between packets, but at the same time, quite a few processing actions require processing the association information between packets, such as counting, rate limiting, etc. Since the network processor is a multi-core architecture, each core processes different data packets, and the processing information shared between packets is generally placed in a shared storage unit (such as on-chip SRAM or off-chip DDR). Then, in the processing of the associated data between packets, a process of "data reading - modification and update - write-back" will occur. During the modification and update process, multiple cores may modify the same data, so atomicity guarantee of the data is required, such as locking, protection, etc.

[0003] Due to the existence of atomic locks between multiple processing engines in the prior art, they have to wait for other processing engines to finish their operations before they can continue processing. The intermediate waiting time adds multiple bus transmission and interaction times, resulting in low efficiency. Summary of the Invention

[0004] In view of this, embodiments of the present application provide a method and system for processing associated data of data packets to eliminate or improve one or more defects existing in the prior art.

[0005] The first aspect of the present application provides a method for processing associated data of data packets executed by a data packet processing engine, the method including:

[0006] Classifying the association data processing instructions generated by itself for the target data packet to obtain target association data processing instructions;

[0007] Sending the target association data processing instructions to the corresponding data processing engine, so that the data processing engine converts the target association data processing instructions into the target address of the target program segment included in the data processing engine, and executes the target program segment to read, update, or write back the target association data in the storage management engine.

[0008] In some embodiments of the present application, before classifying the association data processing instructions generated by itself for the target data packet to obtain target association data processing instructions, it further includes:

[0009] Receiving the target data packet;

[0010] Process the target data packet to generate the associated data processing instruction.

[0011] In some embodiments of the present application, after executing the target program segment to read, update, or write back the target associated data in the storage management engine, it further includes:

[0012] Send the target data packet to the packet cache unit, and based on the packet cache unit, send the target data packet to the data packet forwarding engine.

[0013] In some embodiments of the present application, classifying the associated data processing instruction generated by itself for the target data packet to obtain the target associated data processing instruction includes:

[0014] Based on the instruction type and task identifier of the associated data processing instruction, divide the associated data processing instruction into the corresponding data processing engine to obtain the target associated data processing instruction.

[0015] In some embodiments of the present application, it further includes:

[0016] If the target associated data processing instruction is a read-only instruction, access the storage management engine based on the read-only instruction to obtain the target associated data.

[0017] In some embodiments of the present application, it further includes:

[0018] If the target associated data processing instruction is a write-only instruction, store the target associated data in the storage management engine.

[0019] The second aspect of the present application provides a method for processing packet associated data executed by a data processing engine, the method including:

[0020] Convert the received target associated data processing instruction into the target address of the target program segment included in itself; wherein, the target associated data processing instruction is obtained by classifying the associated data processing instruction generated by the data packet processing engine for the target data packet;

[0021] Execute the target program segment to read, update, or write back the target associated data in the storage management engine.

[0022] In some embodiments of the present application, when the target associated data processing instruction is a preset target instruction, after executing the target program segment to read, update, or write back the target associated data in the storage management engine, it further includes:

[0023] Return the execution result of the target instruction to the data packet processing engine corresponding to the target instruction.

[0024] The third aspect of the present application provides a data packet associated data processing system, which includes: a data packet processing engine and a data processing engine connected by communication;

[0025] The data packet processing engine is used to execute the data packet associated data processing method described in the first aspect;

[0026] The data processing engine is used to execute the data packet associated data processing method described in the second aspect. The fourth aspect of the present application provides a system-on-chip, including a memory, a processor, an on-chip interconnection network, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the data packet associated data processing method described in the foregoing first aspect, or implements the data packet associated data processing method described in the foregoing second aspect.

[0027] The fifth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the data packet associated data processing method described in the foregoing first aspect, or implements the data packet associated data processing method described in the foregoing second aspect.

[0028] The present application provides a data packet associated data processing method and system. The method includes: classifying the associated data processing instructions for the target data packet generated by itself to obtain target associated data processing instructions; sending the target associated data processing instructions to the corresponding data processing engine, so that the data processing engine converts the target associated data processing instructions into the target address of the target program segment included in the data processing engine, and executes the target program segment to read, update, or write back the target associated data in the storage management engine. The present application can effectively ensure the atomicity of data packet associated data processing, and at the same time can effectively improve the concurrent working efficiency of the data processing engine.

[0029] The additional advantages, objectives, and features of the present application will be partially described below, and will become partially apparent to those of ordinary skill in the art after studying the following text, or can be learned from the practice of the present application. The objectives and other advantages of the present application can be achieved and obtained by the structures specifically pointed out in the specification and the drawings.

[0030] Those skilled in the art will understand that the objectives and advantages that can be achieved by the present application are not limited to the above specific descriptions, and the above and other objectives that the present application can achieve will be more clearly understood according to the following detailed description. Description of the Drawings

[0031] The accompanying drawings described herein are used to provide a further understanding of the present application, form a part of the present application, and do not limit the present application. The components in the drawings are not drawn to scale, but are only for showing the principles of the present application. For the convenience of showing and describing some parts of the present application, the corresponding parts in the drawings may be enlarged, that is, they may become larger relative to other components in the exemplary device actually manufactured according to the present application. In the drawings:

[0032] Figure 1 It is a schematic flowchart of a data packet association data processing method in an embodiment of the present application.

[0033] Figure 2 It is a schematic structural diagram of a data packet association data processing device in another embodiment of the present application.

[0034] Figure 3 It is a schematic architecture diagram of a network processor in another embodiment of the present application.

[0035] Figure 4 It is a schematic diagram of the data flow direction of a network processor in another embodiment of the present application. Detailed implementation manners

[0036] To make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below in conjunction with the implementation manners and the accompanying drawings. Herein, the illustrative implementation manners of the present application and their descriptions are used to explain the present application, but do not limit the present application.

[0037] Herein, it also needs to be noted that in order to avoid obscuring the present application due to unnecessary details, only the structures and / or processing steps closely related to the solution according to the present application are shown in the drawings, and other details less related to the present application are omitted.

[0038] It should be emphasized that the term "including / comprising" when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0039] Herein, it also needs to be noted that if not otherwise specified, the term "connection" in this document can not only refer to a direct connection, but also represent an indirect connection with an intermediate.

[0040] In the following, embodiments of the present application will be described with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0041] Specifically, it will be described in detail through the following embodiments.

[0042] An embodiment of the present application provides a method for processing packet - associated data that can be executed by a packet processing engine. Refer to Figure 1 , the method for processing packet - associated data specifically includes the following content:

[0043] Step 110: Classify the association data processing instructions for the target packet generated by itself to obtain the target association data processing instructions.

[0044] Step 120: Send the target association data processing instructions to the corresponding data processing engine, so that the data processing engine converts the target association data processing instructions into the target address of the target program segment included in the data processing engine, and executes the target program segment to read, update or write back the target association data in the storage management engine.

[0045] Specifically, the packet processing engine (i.e., Figure 3 or Figure 4 the PPE in) first classifies the association data processing instructions for the target packet generated by itself to obtain the target association data processing instructions. Then, it sends the target association data processing instructions to the corresponding data processing engine (i.e., Figure 3 or Figure 4 the DPE in), so that the data processing engine converts the target association data processing instructions into the target address of the target program segment (i.e., Figure 4 the Program in), and then executes the target program segment to read, update or write back the target association data in the storage management engine (i.e., Figure 3 or Figure 4 the Uniform Memory Managerment Engine in), thereby effectively ensuring the atomicity of packet - associated data processing and effectively improving the concurrent working efficiency of the data processing engine.

[0046] Among them, the data processing engine (DPE) is composed of units such as an input - output buffer, a program memory, an ALU unit, and a GPR unit (general - purpose register). Intermediate data related to target packet processing is forwarded or output through the bus in a hardware - scheduled or software - scheduled manner among PPEs.

[0047] In addition, refer to Figure 3 , the present application provides a network processor (i.e., Figure 3 the Network Processor in), and this network processor system includes two processing planes. One is the packet processing plane (i.e., Figure 3 the Packet Processing Plane in), and the other is the storage processing plane (i.e., Figure 3(in the Memory Processing Plane). The packet processing plane consists of packet input / output interfaces (i.e., Input and Output in Figure 3 ), packet storage units (i.e., PacketBuffer in Figure 3 ), packet processing engines, and buses (i.e., Bus in Figure 3 ), etc. The storage processing plane consists of storage media such as DDR, on-chip RAM, and TCAM, storage management engines, and data processing engines. The packet processing plane and the storage processing plane are connected by a bus. Figure 3 (Input and Output in Figure 3 ), packet storage units (i.e., PacketBuffer in Figure 3 ), Figure 3 (PacketBuffer in Figure 3 ), packet processing engines, and buses (i.e., Bus in Figure 3 ), etc. Figure 3 (Bus in Figure 3 ), etc. The storage processing plane consists of storage media such as DDR, on-chip RAM, and TCAM, storage management engines, and data processing engines. The packet processing plane and the storage processing plane are connected by a bus. Figure 3 (DirectAccess in Figure 3 ) means directly accessing the encapsulated memory data structure without going through the DPE; Table Structures: are table structures, that is, data structures that encapsulate memory data, enabling visitors (PPE / DPE) not to care about physical memory.

[0048] Among them, both the packet processing engine and the data processing engine support user-defined programming, thus effectively improving the flexibility and scalability of the network processor system.

[0049] It should be noted that both PPE and DPE are components of the on-chip integrated system; in this solution, the network processor collaboratively processes the streaming data entering the system by comprehensively invoking PPE and DPE, thus effectively improving the efficiency of network data processing. This integrated system can be implemented on a single chip or on multiple chips integrated together through packaging.

[0050] Among them, this on-chip integrated system integrates a general-purpose processor, a dedicated processing engine (DSA), and on-chip memories (such as SRAM, DDR, etc., volatile memories or non-volatile memories such as Flash). These units can be integrated on a single die or integrated into a chip through packaging technology on different silicon wafers. The dedicated processing engine can also be a lightweight general-purpose processor (such as riscv).

[0051] In order to effectively obtain the associated data processing instructions, before step 110, it also includes:

[0052] Receiving the target packet;

[0053] Processing the target packet to generate the associated data processing instructions.

[0054] Specifically, after the packet processing engine receives the target packet, it performs corresponding parsing and processing on the target packet, and then obtains the associated data processing instructions, thereby effectively obtaining the associated data instructions.

[0055] To avoid the backlog of tasks in the data packet processing engine and thus improve the working efficiency of the data packet engine, after step 120, the following steps are also included:

[0056] Send the target data packet to the packet buffer unit, and based on this packet buffer unit, send the target data packet to the data packet forwarding engine.

[0057] Specifically, the data packet data engine sends the target data packet to the packet buffer unit (i.e., the Packet Buffer in Figure 3 or Figure 4 ), and based on this packet buffer unit, sends the target data packet to the data packet forwarding engine (i.e., other data packet processing engines communicatively connected to the data packet processing engine, such as the PPE in Figure 3 or Figure 4 ) to further process the target data packet, so as to avoid the backlog of tasks in the data packet processing engine and thus improve the working efficiency of the data packet engine.

[0058] To further improve the concurrent working efficiency of the data processing engine, step 110 includes:

[0059] Divide the associated data processing instruction into the corresponding data processing engine based on the instruction type and task identifier of the associated data processing instruction to obtain the target associated data processing instruction.

[0060] Specifically, the data packet processing engine calls the bus unit to divide the associated data processing instruction into the corresponding data processing engine based on the instruction type and task identifier of the associated data processing instruction to obtain the target associated data processing instruction, so as to further improve the concurrent working efficiency of the data processing engine.

[0061] To improve the efficiency of associated data processing in the case of read-only instructions or write-only instructions, the foregoing data packet associated data processing method further includes:

[0062] If the target associated data processing instruction is a read-only instruction or a write-only instruction, access the storage management engine based on the read-only instruction to obtain the target associated data or store the target associated data in the storage management engine based on the write-only instruction.

[0063] Specifically, refer to Figure 3The instructions in it are transmitted by the Bus to the Uniform Memory Managerment Engine. If the target associated data processing instruction is a read-only instruction or a write-only instruction, the packet processing engine directly accesses the storage management engine based on the read-only instruction to obtain the target associated data, or directly stores the target associated data into the storage management engine based on the write-only instruction, so as to effectively improve the efficiency of associated data processing in the case of read-only instructions or write-only instructions.

[0064] An embodiment of the present application provides a method for processing packet associated data that can be executed by a data processing engine. Refer to Figure 2 , and the method for processing packet associated data specifically includes the following contents:

[0065] Step 210: Convert the received target associated data processing instruction into the target address of the target program segment included in itself; wherein, the target associated data processing instruction is obtained by classifying the associated data processing instruction for the target packet generated by the packet processing engine.

[0066] Step 220: Execute the target program segment to read in, update or write back the target associated data in the storage management engine.

[0067] Specifically, the data processing engine first converts the received target associated data processing instruction into the target address of the target program segment included in itself; then executes the target program segment to read in, update or write back the target associated data in the storage management engine, so as to effectively ensure the atomicity of packet associated data processing and effectively improve the concurrent working efficiency of the data processing engine.

[0068] Among them, the target associated data processing instruction is obtained by classifying the associated data processing instruction for the target packet generated by the packet processing engine.

[0069] In order to effectively limit the traffic of the packet, after step 220, it further includes:

[0070] When the target associated data processing instruction is a preset target instruction, return the execution result of the target instruction to the packet processing engine corresponding to the target instruction.

[0071] Specifically, when the target associated data processing instruction is a preset target instruction, the data processing engine returns the execution result of the preset target instruction to the packet processing engine corresponding to the target instruction, so as to effectively limit the traffic of the packet.

[0072] Since the data processing engine (DPE) is a programmable structure, various service forms can be constructed through certain program configurations. Some examples are as follows:

[0073] Traffic rate limiting: The PPE sends traffic rate limiting instructions and parameters to the DPE. The DPE is controlled by a program, reads historical statistical data in the buffer (i.e., reads data in the buffer through the functions encapsulated by the storage management engine), performs calculations, returns the calculation results to the PPE, and writes the updated values into the buffer at the same time.

[0074] Linked list structure construction: The linked list structure is maintained and operated by the DPE in the buffer. The PPE sends linked list operation instructions (such as node insertion, deletion, etc.) to the DPE. The DPE is controlled by a program, reads the relevant information of the node in the buffer, performs modification operations, and returns the results (such as the newly applied node number) to the PPE.

[0075] Timer operation: The DPE maintains timer data in the buffer. The DPE program runs independently. When the PPE sends timer operation instructions and timer instance parameters to the DPE, the DPE sorts the timer instances, writes the request parameters into the buffer, and real-time detects the expired timer data. When it detects that a timer reaches the predetermined time, it returns to the specified PPE according to the recorded information.

[0076] An embodiment of the present application further provides a packet associated data processing system, which includes: a packet processing engine and a data processing engine that are communicatively connected;

[0077] The packet processing engine is used to execute the foregoing first packet associated data processing method;

[0078] The data processing engine is used to execute the foregoing second packet associated data processing method.

[0079] The present application provides a packet associated data processing method and system. The method includes: classifying the association data processing instructions for the target packet generated by itself to obtain target association data processing instructions; sending the target association data processing instructions to the corresponding data processing engine, so that the data processing engine converts the target association data processing instructions into the target address of the target program segment included in the data processing engine, and executes the target program segment to read in, update, or write back the target association data in the storage management engine. The present application can effectively ensure the atomicity of packet associated data processing and can effectively improve the concurrent working efficiency of the data processing engine.

[0080] Embodiments of the present application further provide a system-on-chip, such as a DPU chip. The system-on-chip may include a processor, a memory, a receiver, an on-chip interconnection network, and a transmitter. The processor is configured to execute the packet association data processing method mentioned in the above embodiments. The processor and the memory may be connected through a bus or other means. Taking the connection through the bus as an example, the receiver may be connected to the processor and the memory through on-chip interconnection or chip bonding, etc.

[0081] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. chips, or a combination of the above chips or integrated circuit IP instances. The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the packet association data processing method in the embodiments of the present application. The processor executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory, that is, implements the packet association data processing method in the above method embodiments.

[0082] The memory may include a program storage area and a data storage area. Among them, the program storage area can store application programs required for at least one function; the data storage area can store data created by the processor, etc. In addition, the memory may include high-speed random access memory and may also include non-transitory memory. In some embodiments, the memory may optionally include a memory remotely provided relative to the processor, and these remote memories may be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and combinations thereof.

[0083] The one or more modules are stored in the memory and, when executed by the processor, execute the packet association data processing method in the embodiments.

[0084] The embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the foregoing data packet association data processing method are implemented. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, floppy disk, hard disk, removable storage disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0085] Those of ordinary skill in the art should understand that the various exemplary components, systems, and methods described in connection with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Specifically, whether to implement in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments used to execute the required tasks. The programs or code segments can be stored in a machine-readable medium, or transmitted through a data signal carried in a carrier wave on a transmission medium or a communication link.

[0086] It should be clear that the present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0087] In the present application, the features described and / or illustrated for one embodiment can be used in the same or similar manner in one or more other embodiments, and / or combined with the features of other embodiments or replace the features of other embodiments.

[0088] The above are only the preferred embodiments of the present application and are not used to limit the present application. For those skilled in the art, the embodiments of the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A method for processing packet association data, characterized in that, This method is executed by a data packet processing engine and a data processing engine, and multiple said data packet processing engines and multiple data processing engines are all components in a system-on-chip; The system-on-chip is implemented by a DPU chip; The method includes: The data packet processing engine classifies the associated data processing instructions generated by itself for a target data packet to obtain target associated data processing instructions; The data packet processing engine issues the target associated data processing instructions to the corresponding data processing engine, so that the data processing engine converts the target associated data processing instructions into the target address of the target program segment included in the data processing engine, and executes the target program segment to read in, update or write back the target associated data in the storage management engine; The data packet processing engine sends the target data packet to the packet buffer unit, and based on this packet buffer unit, sends the target data packet to other data packet processing engines communicatively connected to this data packet processing engine, so that other data packet processing engines process the target data packet; The method further includes: Traffic rate limiting: The data packet processing engine sends a traffic rate limiting instruction and parameters to the data processing engine. The data processing engine is controlled by a program, reads historical statistical data in the buffer area through the function encapsulated by the storage management engine, performs calculations, and returns the calculation result to the data packet processing engine, and at the same time writes the updated value into the buffer area; Linked list structure construction: The linked list structure is maintained and operated by the data processing engine in the buffer area. The data packet processing engine sends a linked list operation instruction to the data processing engine. The data processing engine is controlled by a program, reads node information in the buffer area, performs modification operations, and returns the result to the data packet processing engine; Timer operation: The data processing engine maintains timer data in the buffer area. The data processing engine program runs independently. When the data packet processing engine sends a timing operation instruction and timer instance parameters to the data processing engine, the data processing engine sorts the timer instances, writes the request parameters into the buffer area, and continuously detects the timer data that has reached the scheduled time. When it detects that a timer has reached the scheduled time, it returns to the specified data packet processing engine according to the recorded information.

2. The data packet association data processing method according to claim 1, wherein Before the data packet processing engine classifies the associated data processing instructions generated by itself for a target data packet to obtain target associated data processing instructions, it further includes: Receiving the target data packet; Processing the target data packet to generate the associated data processing instructions.

3. The data packet association data processing method according to claim 1, wherein The data packet processing engine classifies the associated data processing instructions generated by itself for a target data packet to obtain target associated data processing instructions, including: Dividing the associated data processing instructions into the corresponding data processing engines based on the instruction type and task identifier of the associated data processing instructions to obtain the target associated data processing instructions.

4. The data packet association data processing method according to claim 1, characterized in that, It further includes: If the target associated data processing instruction is a read-only instruction, access the storage management engine based on the read-only instruction to obtain the target associated data.

5. The data packet association data processing method according to claim 1, wherein It further includes: If the target associated data processing instruction is a write-only instruction, store the target associated data in the storage management engine.

6. A data packet associated data processing system, characterized in that, Including: A packet processing engine and a data processing engine connected by communication; The packet processing engine and the data processing engine are used to execute the packet associated data processing method according to any one of claims 1-5.

7. An on-chip integrated system, comprising a memory, a processor, an on-chip interconnection network, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the packet associated data processing method according to any one of claims 1 to 5.

Images