CN116095195B - A method and device for generating industrial Ethernet data packets based on a localized platform - Google Patents

Abstract

The invention discloses a method and device for generating industrial Ethernet data packets based on a localized platform. The method can cover the continuous generation of typical industrial Ethernet protocol characteristic data packets, and is useful for industrial Ethernet traffic analysis research, industrial Ethernet security analysis, Test verification provides direct support and has a good application prospect. At the same time, the time stamp counter (TSC) is used for high-resolution time measurement to solve the problem of inability to accurately record a given time point caused by the low timing accuracy of the operating system.

Description

A method and device for generating industrial Ethernet data packets based on a localized platform

technical field

The invention relates to the technical field of network testing, in particular to a method and device for generating industrial Ethernet data packets based on a localized platform that can cover the continuous generation of characteristic data packets of typical industrial Ethernet protocols.

Background technique

Network security is information security on the network, which means that the hardware, software and data in the network system are protected from being damaged, changed, or leaked due to accidental or malicious reasons, and the system runs continuously and reliably. Network services are not interrupted.

With the deep integration of national informatization and industrialization, a large number of informatization, intelligent products and advanced technical means have been applied to various industrial control systems, breaking the original physical isolation of traditional industrial enterprise networks. When various technologies are introduced, information technology security risks are also introduced, which brings great challenges to the security protection of industrial Ethernet. For this reason, a large number of domestic information security manufacturers have also increased their investment and launched many industrial Ethernet security devices and software.

After the development of industrial Ethernet security equipment and software is completed, its performance needs to be verified. In order to ensure the accuracy of performance verification, it is usually necessary to connect industrial Ethernet security equipment and software to the corresponding industrial Ethernet in normal operation. However, because industrial Ethernet is a high-value network, any interruption of operation will directly affect the production rhythm of the enterprise. Conventional security equipment testing and verification methods will mostly monitor and analyze network data. Ethernet may affect the normal timing of Ethernet, which will affect the real-time and timing of normal data in the industrial network, and have unpredictable effects on the operation of industrial equipment, and the lack of verification of industrial Ethernet security equipment cannot reflect its effectiveness. sex.

In order to solve the above problems, a network data packet generator is proposed. The network data packet generator (also known as a network tester, which can be referred to as a data packet generator) generates a specific network data packet and sends the network data packet to the network to be tested. In the security device, it receives the data packet returned by the network device to be tested to realize functions such as performance test, function test and network connectivity test. However, the network data packet generators in the prior art generally have a single function. Due to the large amount of data in the simulated industrial Ethernet, the traditional network data packet generators still have slow data packet generation speed, data transmission interruption and card failure. broken problem.

In addition, most of the network data packet generation devices currently used in China are based on foreign hardware platforms and developed by foreign engineers. In view of the current international situation, their data security and technical availability cannot be guaranteed. With the continuous development of localization technology, the future network packet generation device based on the localization platform also needs in-depth research. Therefore, there is an urgent need for a device based on a localized platform that can simulate and generate full coverage of industrial Ethernet feature data to provide basic support for the development of industrial Ethernet security technology and industrial Ethernet performance testing.

Contents of the invention

In view of the above problems, the present invention provides a method and device for generating industrial Ethernet data packets based on a localized platform for overcoming the above problems or at least partially solving the above problems. It can cover typical industrial Ethernet protocol characteristic data packets and continuously generate, providing direct support for industrial Ethernet traffic analysis research, industrial Ethernet security analysis, test verification, and has a good application prospect.

The present invention provides following scheme:

A method for generating industrial Ethernet data packets based on a localization platform, comprising:

Obtain several data files stored in the file system, each of which includes industrial Ethernet data to be simulated in a data frame type;

storing the data frame loads contained in the plurality of data files in a plurality of ring buffers in a one-to-one correspondence;

Using the multi-core parallel serialization model to serialize the data frame loads in the ring buffer one by one to obtain a number of serialization records, and push some of the serialization records into a multi-producer and multi-consumer lock-free queue; The multi-core parallel serialization model includes several first parallel worker threads, and the serialization record includes a target serialization format;

Using the parallel sending model to take out some of the serialized records in the multi-producer and multi-consumer lock-free queue, and send them out through the network card in a direct memory access mode; the parallel sending model includes several second parallel jobs thread.

Preferably: the target serialization format includes a length field, a timestamp field, and a data payload field arranged in sequence from left to right; the length field is used to indicate the length of the data packet load, and the timestamp field is used to indicate the future of the data packet At the time of sending, the data payload field is used to indicate the data payload directly sent when the data packet is sent.

Preferably: the content stored in the data file includes all data frame loads above the physical layer in the OSI seven-layer protocol in the industrial Ethernet data to be simulated of the same data frame type.

Preferably: using a data creation thread to create and obtain several ring buffers, the number of several ring buffers is the same as the number of data files, and the number of nodes contained in each ring buffer is the same as that of each ring buffer The number of data contained in the data file is the same, and the content of the node includes the content of the data load, the length of the data load, and the speed of data transmission.

Preferably: each of the first parallel worker threads successively transfers the data in the ring buffer according to the serial order of the ring buffer linked list at a consumption speed that is multiple times that of the multi-producer and multi-consumer lock-free queue. The data frame load is taken out and serialized to obtain serialized records.

Preferably: the multi-core parallel serialization model single-threaded working method includes:

After determining the empty semaphore of the multi-producer and multi-consumer lock-free queue, fetch the data load content, data load length and data transmission speed related information from the corresponding ring buffer;

Allocating memory from a memory pool to create the serialization record and filling the data length into the serialization record, the memory pool being created by the data creation thread.

Preferably: determining that the serialized record is the first serialized record generated by the first parallel worker thread;

Obtaining the system timestamp counter value and calculating the first sending timestamp of the first serialized record and storing it in a global variable as a benchmark reference point for subsequent generation of all the serialized records;

Calculate and obtain the future sending time of the data frame of the first serialized record according to the future sending timestamp of the data frame of the first serialized record, the value of the system timestamp counter, and the first sending timestamp, and store the data Fill in the serialization record at the time when the frame is sent in the future;

Copy the data frame payload content to the serialized record;

Push the serialized record into a multi-producer and multi-consumer lock-free queue.

Preferably: using the parallel sending model to take out several of the serialized records in the multi-producer and multi-consumer lock-free queue, and send them out through the network card in the form of direct memory access, including:

Confirm that this sending is the first sending;

Record the second sending time stamp of the data, and use the second sending time stamp and the system time stamp counter value as a benchmark reference point for all subsequent single-thread sending data.

Preferably: the file system is arranged on a Feiteng platform based on localization, and the network card is a DMA network card based on the Feiteng platform.

An industrial Ethernet data packet generation device based on a localized platform, comprising:

The data file acquisition unit acquires several data files stored in the file system, each of which includes industrial Ethernet data to be simulated in a data frame type;

The cache unit is used to store the data frame loads contained in the plurality of data files in a plurality of ring buffers in a one-to-one correspondence; the data creation thread is used to create and obtain a plurality of the ring buffers, and the number of the plurality of ring buffers is The same as the number of several data files, the number of nodes contained in each ring buffer is the same as the number of data contained in each data file, and the content of the nodes includes data load content, data load length, data transmission speed;

The serialization recording unit is used to serialize the data frame loads in the ring buffer one by one by using the multi-core parallel serialization model to obtain a number of serialization records, and push the number of serialization records into the multi-producer multiple Consumer lock-free queue; the multi-core parallel serialization model includes several first parallel worker threads, the serialization records include the target serialization format, each of the first parallel worker threads is connected in series according to the ring buffer linked list Sequentially take out the data frame load in the ring buffer at a consumption speed that is multiple times that of the multi-producer multi-consumer lock-free queue and serialize to obtain serialized records; the multi-core parallel serialization model is single-threaded Working methods include:

After determining the empty semaphore of the multi-producer multi-consumer lock-free queue, take out the load content, data load length, and data transmission speed related information from the corresponding ring buffer;

Allocating memory from a memory pool to create the serialized record and filling the data length into the serialized record, the memory pool being created and obtained by the data creation thread;

A sending unit, configured to take out several of the serialized records in the multi-producer and multi-consumer lock-free queue by using a parallel sending model, and send them outward through the network card in a direct memory access mode; the parallel sending model includes A number of second parallel worker threads. According to the specific embodiments provided by the invention, the invention discloses the following technical effects:

The embodiment of the present application provides a method and device for generating industrial Ethernet data packets based on a localized platform, which can cover the continuous generation of characteristic data packets of typical industrial Ethernet protocols, and is used for industrial Ethernet traffic analysis research, industrial Ethernet security analysis, Test verification provides direct support and has a good application prospect. At the same time, the time stamp counter (TSC) is used for high-resolution time measurement to solve the problem of inability to accurately record a given time point caused by the low timing accuracy of the operating system.

The serialization thread uses the first sending timestamp Tg generated by TSC combination as the time reference for subsequent production records, and generates all sending data records with the same time reference across multiple production threads, which solves the consistency problem of generating record time across multiple threads .

In the sending single thread, TSC is used in combination with the first sending time stamp T0 as the time reference for subsequent sending records, and the sending time point is expressed as a relative quantity, which solves the problem of time synchronization of all sending records in a single thread, and solves the problem of previous traffic The generator cannot continuously send network traffic with a variable rate, which has higher application value and covers a wider range of industrial Ethernet real network traffic application scenarios.

By storing the generated data, the size of the sent traffic and the sending rate in the file system based on the localization platform, and adopting an editable mode with the data content fully open, the data packet generator can cover the full protocol based on Ethernet, the full packet length, Full data content, full data frame mixing ratio, and full flow size customize the application scenarios for industrial Ethernet data frame generation.

Of course, any product implementing the present invention does not necessarily need to achieve all the above-mentioned advantages at the same time.

Description of drawings

In order to illustrate the embodiments of the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the drawings required in the embodiments. Apparently, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings according to these drawings without creative efforts.

Fig. 1 is a flow chart of a method for generating industrial Ethernet data packets based on a localization platform provided by an embodiment of the present invention;

Fig. 2 is a schematic diagram of the architecture of the data generator provided by the embodiment of the present invention;

Fig. 3 is a schematic diagram of a serialized record format provided by an embodiment of the present invention;

FIG. 4 is a single-thread workflow diagram of the multi-core parallel serialization model provided by the embodiment of the present invention;

Fig. 5 is a working flowchart of the parallel transmission model provided by the embodiment of the present invention;

Fig. 6 is a schematic diagram of the Feiteng hardware platform provided by the embodiment of the present invention;

Fig. 7 is a schematic diagram of an industrial Ethernet data packet generating device based on a localization platform provided by an embodiment of the present invention;

Fig. 8 is a schematic diagram of an industrial Ethernet data packet generation device based on a localized platform provided by an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Apparently, the described embodiments are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention belong to the protection scope of the present invention.

Referring to Fig. 1, a kind of industrial Ethernet packet generation method based on the localization platform provided by the embodiment of the present invention, as shown in Fig. 1, the method may include:

S101: Obtain several data files stored in the file system, each of which includes a data frame type of industrial Ethernet data to be simulated; specifically, the content stored in the data file includes the same data frame type All data frame loads above the physical layer in the OSI seven-layer protocol in the industrial Ethernet data to be simulated.

S102: Store the data frame loads contained in the plurality of data files in a plurality of ring buffers in one-to-one correspondence; specifically, use the data creation thread to create and obtain a plurality of the ring buffers, and the number of the plurality of ring buffers The same as the number of several data files, the number of nodes contained in each ring buffer is the same as the number of data contained in each data file, and the content of the nodes includes data load content, data load length, Data sending speed.

S103: Use the multi-core parallel serialization model to serialize the data frame loads in the ring buffer one by one to obtain a number of serialized records, and push the serialized records into a multi-producer and multi-consumer lock-free queue ; The multi-core parallel serialization model includes several first parallel worker threads, and the serialization record includes a target serialization format; specifically, the target serialization format includes a length field, a timestamp field and data payload field; the length field is used to indicate the length of the data packet payload, the timestamp field is used to indicate the future sending time of the data packet, and the data payload field is used to indicate the data payload directly sent when the data packet is sent.

In practical applications, each of the first parallel worker threads sequentially converts the ring buffer into the ring buffer at a consumption speed that is multiple times that of the multi-producer multi-consumer lock-free queue according to the serial order of the ring buffer linked list. The data frame load within is fetched and serialized to obtain serialized records.

The single-thread working method of the multi-core parallel serialization model includes:

After determining the empty semaphore of the multi-producer and multi-consumer lock-free queue, fetch the data load content, data load length and data transmission speed related information from the corresponding ring buffer;

Allocating memory from a memory pool to create the serialization record and filling the data length into the serialization record, the memory pool being created by the data creation thread.

determining that the serialization record is the first serialization record generated by the first parallel worker thread;

Obtaining the system timestamp counter value and calculating the first sending timestamp of the first serialized record and storing it in a global variable as a benchmark reference point for subsequent generation of all the serialized records;

Calculate and obtain the future sending time of the data frame of the first serialized record according to the future sending timestamp of the data frame of the first serialized record, the value of the system timestamp counter, and the first sending timestamp, and store the data Fill in the serialization record at the time when the frame is sent in the future;

Copy the data frame payload content to the serialized record;

Push the serialized record into a multi-producer and multi-consumer lock-free queue.

S104: Use the parallel sending model to take out several of the serialized records in the multi-producer and multi-consumer lock-free queue, and use direct memory access to send them out through the network card; the parallel sending model includes several second Parallel worker threads. Specifically, it is determined that this sending is the first sending;

Record the second sending time stamp of the data, and use the second sending time stamp and the system time stamp counter value as a benchmark reference point for all subsequent single-thread sending data.

In order to make the industrial Ethernet data packet generation method based on the localization platform provided by the embodiment of the present application can be used based on the localization software, the embodiment of the present application can also provide that the file system is arranged in the Feiteng platform based on the localization, so The network card mentioned above is a DMA network card based on the Phytium platform.

The industrial Ethernet data packet generation method based on the localization platform provided by the embodiment of the present application can simulate and generate full coverage of industrial Ethernet characteristic data. When testing equipment, it is not necessary to directly connect the security device to the already running industrial Ethernet. The method provided by the embodiment of the present application can quickly generate data packets for simulating related industrial Ethernet, and realize high-simulation testing of network security equipment without destroying the normal real-time performance and timing of the already running industrial Ethernet. Provide basic support for the development of industrial Ethernet security technology and industrial Ethernet performance testing.

It can be understood that the industrial Ethernet data packet generation method based on the localization platform provided by the embodiment of the present application can be used in the network data packet generator, the network data packet generator (also known as the network tester, can be referred to as the data packet The generator) generates a specific network data packet, sends the network data packet to the network device to be tested, and receives the data packet returned by the network device to be tested to realize functions such as performance test, function test and network connectivity test. Packet generators play an important role in network research, development, and operation and maintenance. Network researchers can use the data packet generator to test the performance and functions of the network prototype system and find system bottlenecks; network operation and maintenance personnel can use data packets to debug network faults and test network service quality. Packet generator is an indispensable network tool with important socio-economic benefits and academic research value.

The following uses the method provided by the embodiment of the present application based on the Feiteng platform as an example to describe in detail.

The embodiment of this application can adopt the national production Feiteng hardware platform design, as shown in Figure 6, the FT-2000/4 general-purpose computing processor chip adopted by the hardware platform integrates 4 processor cores FTC663 independently developed by Feiteng, and is compatible with 64-bit ARMv8 Instruction set, 16nm process, main frequency up to 3.0GHz, two 72-bit (support ECC) DDR4-3200 storage interfaces, two 16lane (each can be split into 2 x8), two 1lane PCIe3.0, support up to 6 PCIe3.0 ports and two Gigabit Ethernet ports. FCBGA package, the number of pins is 1144, the package size is 35mmx35mm, and the maximum power consumption is 10W.

Figure 2 shows the architecture of the packet generator. The generator consists of a data production module, a multi-core parallel serialization model and a multi-core parallel serialization model. The following four components are described in detail.

Data production module: consists of data files, data creation threads and ring buffers.

The data files are stored in the data files of the file system based on the localized Feiteng platform, each data file represents a type of industrial Ethernet data frame, and the content stored in the data file is all data frames above the physical layer in the OSI seven-layer protocol load. When different types of industrial Ethernet data need to be generated, the data of different frame types are stored in different files. When it is necessary to generate industrial Ethernet data frames of the same type with different lengths and different contents, store the data frame load in the same file.

The data creation thread is responsible for creating a memory pool for the MPMC queue to allocate memory when the generator is initialized. This design can greatly speed up the creation of MPMC queue nodes. Also create a ring buffer, read the data file content and copy it to the ring buffer.

The ring buffer stores data frame payloads extracted from data files, consisting of data blocks based on a doubly linked list. The ring buffer is created by the data creation thread. The number depends on the number of data files. The number of nodes in each ring buffer depends on the number of data in each data file. The content of the nodes includes the content of the data load, the length of the data load, and the speed of sending data .

Multi-core parallel serialization model: composed of multiple first parallel worker threads, the number of first parallel worker threads depends on the number of ring buffers. Each thread sequentially fetches the data according to the concatenation order of the ring buffer linked list at a consumption speed many times faster than that of the MPMC lock-free queue, and then serializes it into a record according to the target serialization format shown in Figure 3, and pushes it into the MPMC Lock-free queue. Here, the data is fetched and serialized into records at a speed that is multiple times that of the sending thread, which can ensure that the production rate of the MPMC lock-free queue is always greater than the consumption rate, and will not cause interruption or lag in sending data.

The target serialization format includes a length field, a timestamp field, and a data payload field arranged in sequence from left to right.

length field: 2 bytes, representing the length of the data packet payload.

timestamp field: 8 bytes, the high four bytes are seconds, and the low four bytes are microseconds, compatible with the structtimeval structure under the Linux system, representing the future sending time of the data packet.

data payload field: the actual data payload, the data sent directly when the data packet is sent.

The single-thread workflow of the multi-core parallel serialization model is shown in Figure 4:

Step 1: the global variable MPMC queue length is cleared to zero, wait for the sending thread to complete initialization and inform MPMC by semaphore that the lock-free queue is empty, after receiving the empty semaphore of MPMC lock-free queue, enter step 2, otherwise continue to execute step 1.

Step 2: Take out data, data length and sending speed related information from the corresponding ring buffer, allocate memory from the memory pool to create a serialized record and fill the data length into the serialized record.

Step 3: Detect whether the record is the first record generated by all serialization threads. If it is the first record, first execute step 3-1, and then execute step 3-2. If it is not the first record, then directly execute step 3-2 .

Step 3-1: Obtain the system timestamp counter value TSC (Ch) of the timestamp counter and calculate the first sent timestamp Tg (first sent timestamp) and store it in a global variable as a benchmark reference for subsequent generation of all serialized records point.

Step 3-2: Calculate the future sending time of the data frame according to the future sending time stamp Tn of the data frame in conjunction with the system time stamp counter value TSC (Ch) and the first sending time stamp Tg, sending time T=TSC(Ch)+(Tn- Tg), and fill the sending time into the serialized record.

Step 4: Copy the data frame payload content to the serialized record.

Step 5: Push the serialized record into the MPMC lock-free queue.

Step 6: Detect whether the number of queues created this time reaches the maximum number of queues. If the maximum number of queues is reached, then return to step 1. If the maximum number of queues is not reached, then return to step 2.

A linked list-based lock-free queue created by the serialization thread, each node of the queue contains the serialized records created by the serialization thread. The number of lock-free queues is determined by the number of serialization threads, and the maximum length of the queue is configurable. The consumer of the lock-free data structure is the sending thread, which is acquired and released by the sending thread.

Multi-core parallel serialization model: It is composed of multiple second parallel worker threads, which are responsible for taking out the data in the MPMC queue and sending them out through the Feiteng platform network card by means of direct memory access (DMA). It is a consumer of the lock-free data structure. The workflow of the multi-core parallel serialization model is shown in Figure 5:

Step 1: the number of records sent in this round is cleared, and the semaphore that the lock-free queue is empty is sent to the serialization thread, and the serialization thread is notified to push the data that needs to be sent to the MPMC lock-free queue.

Step 2: detect whether the MPMC lock-free queue is empty, if not empty, then perform step 3, if empty, then sleep for 10ms, and continue to perform step 2.

Step 3: Take out the data in the MPMC lock-free queue head, and the current round of sending frame counts increases by 1.

Step 4: detect whether the nodes in the MPMC lock-free queue to be sent in this round are about to be exhausted, and the exhaustion condition is that the sending record count of the current round reaches the maximum queue length number. If it is exhausted, go to step 4-1, if not, go to step 5.

Step 4-1: The record count of this round is cleared, and the semaphore that the lock-free queue is empty is sent to the serialization thread, and the serialization thread is notified to push the data to be sent in the next round to the MPMC lock-free queue.

Step 5: Extract the data load to be sent and the future sending timestamp Tn of the data from the sending record, and release the memory used by the data record to the memory pool.

Step 6: Detect whether this transmission is the first transmission after the data generator is started, if it is the first transmission, then perform step 6-1, if it is not the first transmission, then perform step 7.

Step 6-1: Record the data sending time stamp as T0 (second sending time stamp), obtain and record the system time stamp counter value TSC(Ch), these two values are used as the benchmark reference point for all subsequent single-thread sending data.

Step 7: Send the time stamp Tn in the future according to the data extracted in step 5, and calculate the theoretical sending time point TSC(Cd)=Tn-T0 with T0 as the reference point, and obtain the current host time stamp counter value TSC(Cn).

Step 8: Calculate the current host time stamp counter value Cn and the system time stamp counter value Ch+calculate the theoretical sending time point Cd size relationship;

If Cn>Ch +Cd, perform step 8-1;

If Cn=>Ch +Cd, execute step 8-2;

If Cn=Ch+Cd, perform step 8-3.

Step 8-1: before the sending time point, skip to step 7 and wait for Cn=>Ch +Cd.

Step 8-2: If the sending time point is exceeded, the data packet violates the rules, the frame is discarded, and the execution is skipped to step 3.

Step 8-3: When the sending time is up, send the data frame to the DMA network card based on the Feiteng platform.

It can be seen that the method provided by the embodiment of the present application uses a time stamp counter (TSC) for high-resolution time measurement to solve the problem that the operating system cannot accurately send at a given time point due to the low timing accuracy of the operating system.

In the sending single thread, TSC is used in combination with the first sending time stamp T0 as the time reference for subsequent sending records, and the sending time point is expressed as a relative quantity, which solves the problem of time synchronization of all sending records in a single thread, and solves the problem of previous traffic The generator cannot continuously send network traffic with a variable rate, which has higher application value and covers a wider range of industrial Ethernet real network traffic application scenarios.

The serialization thread uses the first sending timestamp Tg generated by TSC combination as the time reference for subsequent production records, and generates all sending data records with the same time reference across multiple production threads, which solves the consistency problem of generating record time across multiple threads .

By storing the generated data, the size of the sent traffic and the sending rate in the file system based on the localization platform, and adopting an editable mode with the data content fully open, the data packet generator can cover the full protocol based on Ethernet, the full packet length, Full data content, full data frame mixing ratio, and full flow size customize the application scenarios for industrial Ethernet data frame generation.

Referring to FIG. 7, the embodiment of the present application can also provide a device for generating industrial Ethernet data packets based on a localized platform. As shown in FIG. 7, the device can include:

The data file obtaining unit 701 obtains several data files stored in the file system, each of which includes a data frame type of industrial Ethernet data to be simulated;

The cache unit 702 is used to store the data frame loads contained in the plurality of data files in a plurality of ring buffers in a one-to-one correspondence; use the data creation thread to create and obtain a plurality of the ring buffers, and the number of the ring buffers The number is the same as the number of several data files, the number of nodes included in each ring buffer is the same as the number of data included in each data file, and the content of the nodes includes data load content, data load length , data transmission speed;

The serialization recording unit 703 is used to serialize the data frame loads in the ring buffer one by one by using the multi-core parallel serialization model to obtain a number of serialization records, and push the number of serialization records into the multi-producer Multi-consumer lock-free queue; the multi-core parallel serialization model includes a number of first parallel worker threads, the serialization record includes a target serialization format, and each of the first parallel worker threads follows the ring buffer linked list The series sequence is multiple times faster than the multi-producer multi-consumer lock-free queue, sequentially taking out the data frame load in the ring buffer for serialization to obtain serialized records; the multi-core parallel serialization model unit Thread work methods include:

After determining the empty semaphore of the multi-producer multi-consumer lock-free queue, take out the load content, data load length, and data transmission speed related information from the corresponding ring buffer;

Allocating memory from a memory pool to create the serialized record and filling the data length into the serialized record, the memory pool being created and obtained by the data creation thread;

The sending unit 704 is configured to use a parallel sending model to take out several of the serialized records in the multi-producer and multi-consumer lock-free queue, and send them outward through the network card in a direct memory access mode; the parallel sending model A number of second parallel worker threads are included.

The embodiment of the present application can also provide an industrial Ethernet data packet generation device based on a localization platform, the device includes a processor and a memory:

The memory is used to store program codes and transmit the program codes to the processor;

The processor is configured to execute the steps of the above-mentioned method for generating industrial Ethernet data packets based on a localized platform according to instructions in the program code.

As shown in FIG. 8 , the embodiment of the present application provides an industrial Ethernet data packet generation device based on a localized platform, which may include: a processor 10 , a memory 11 , a communication interface 12 and a communication bus 13 . The processor 10 , the memory 11 , and the communication interface 12 all communicate with each other through the communication bus 13 .

In the embodiment of the present application, the processor 10 may be a central processing unit (Central Processing Unit, CPU), an application-specific integrated circuit, a digital signal processor, a field programmable gate array, or other programmable logic devices.

The processor 10 can call the program stored in the memory 11, specifically, the processor 10 can execute the operations in the embodiment of the method for generating an industrial Ethernet data packet based on a localized platform.

The memory 11 is used to store one or more programs. The programs may include program codes, and the program codes include computer operation instructions. In the embodiment of the present application, the memory 11 stores at least programs for realizing the following functions:

Obtain several data files stored in the file system, each of which includes industrial Ethernet data to be simulated in a data frame type;

storing the data frame loads contained in the plurality of data files in a plurality of ring buffers in a one-to-one correspondence;

Using the multi-core parallel serialization model to serialize the data frame loads in the ring buffer one by one to obtain a number of serialization records, and push some of the serialization records into a multi-producer and multi-consumer lock-free queue; The multi-core parallel serialization model includes several first parallel worker threads, and the serialization record includes a target serialization format;

Using the parallel sending model to take out some of the serialized records in the multi-producer and multi-consumer lock-free queue, and send them out through the network card in a direct memory access mode; the parallel sending model includes several second parallel jobs thread.

In a possible implementation manner, the memory 11 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required by a function (such as file creation function, data reading and writing function) Programs, etc.; the storage data area can store data created during use, such as initialization data.

In addition, the memory 11 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device or other volatile solid-state storage devices.

The communication interface 12 may be an interface of a communication module, and is used for connecting with other devices or systems.

Of course, it should be noted that the structure shown in Figure 8 does not constitute a limitation to the industrial Ethernet data packet generation device based on the localization platform in the embodiment of the application. In practical applications, the industrial Ethernet data packet based on the localization platform The generating device may include more or fewer components than shown in FIG. 8, or combine certain components.

The embodiment of the present application can also provide a computer-readable storage medium, the computer-readable storage medium is used to store program code, and the program code is used to execute the above-mentioned industrial Ethernet data packet generation method based on the localization platform step.

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is a relationship between these entities or operations. There is no such actual relationship or order between them. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

It can be known from the above description of the implementation manners that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the essence of the technical solution of this application or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in storage media, such as ROM/RAM, disk , CD, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments of the present application.

Each embodiment in this specification is described in a progressive manner, the same and similar parts of each embodiment can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to part of the description of the method embodiment. The systems and system embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is It can be located in one place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without creative effort.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present invention are included in the protection scope of the present invention.

Claims (7)

1. A method for generating industrial Ethernet data packets based on a localization platform, characterized in that, comprising: Obtain several data files stored in the file system, each of which includes industrial Ethernet data to be simulated in a data frame type; Store the data frame loads contained in several of the data files in a number of ring buffers in a one-to-one correspondence; use the data creation thread to create and obtain a number of the ring buffers, the number of the number of the ring buffers and the number of the data The number of files is the same, the number of nodes contained in each ring buffer is the same as the number of data contained in each data file, and the content of the nodes includes data load content, data load length, and data transmission speed; Using the multi-core parallel serialization model to serialize the data frame loads in the ring buffer one by one to obtain some serialized records, and push some of the serialized records into a multi-producer and multi-consumer lock-free queue; The multi-core parallel serialization model includes several first parallel worker threads, and the serialization records include the target serialization format; each of the first parallel worker threads is multiple times the The consumption speed of multi-producer and multi-consumer lock-free queue, sequentially take out the data frame load in the ring buffer and serialize to obtain serialized records; the single-threaded working method of the multi-core parallel serialization model includes: After determining the empty semaphore of the multi-producer and multi-consumer lock-free queue, fetch the data load content, data load length and data transmission speed related information from the corresponding ring buffer; Allocating memory from a memory pool to create the serialized record and filling the data length into the serialized record, the memory pool being created and obtained by the data creation thread; Using the parallel sending model to take out some of the serialized records in the multi-producer and multi-consumer lock-free queue, and send them out through the network card in a direct memory access mode; the parallel sending model includes several second parallel jobs thread. 2. the industrial ethernet packet generation method based on the localization platform according to claim 1, is characterized in that, described target serialization format comprises length domain, timestamp domain and data payload domain that are arranged successively from left to right; The length field is used to indicate the length of the data packet payload, the timestamp field is used to indicate the future sending time of the data packet, and the data payload field is used to indicate the data payload directly sent when the data packet is sent. 3. the method for generating industrial Ethernet data packets based on the localization platform according to claim 1, wherein the content stored in the data file includes the OSI data in the industrial Ethernet data to be simulated of the same data frame type All data frame loads above the physical layer in the seven-layer protocol. 4. the industrial Ethernet packet generation method based on the localization platform according to claim 1, is characterized in that, it is determined that the serialization record is the first serialization record that the first parallel worker thread generates; Obtaining the system timestamp counter value and calculating the first sending timestamp of the first serialized record and storing it in a global variable as a benchmark reference point for subsequent generation of all the serialized records; Calculate and obtain the future sending time of the data frame of the first serialized record according to the future sending timestamp of the data frame of the first serialized record, the value of the system timestamp counter, and the first sending timestamp, and store the data Fill in the serialization record at the time when the frame is sent in the future; Copy the data frame payload content to the serialized record; Push the serialized record into a multi-producer and multi-consumer lock-free queue. 5. the industrial ethernet packet generation method based on the localization platform according to claim 4, is characterized in that, described employing parallel transmission model will be described in some described sequences in the multi-producer multi-consumer lock-free queue The records are taken out and sent out through the network card in the way of direct memory access, including: Confirm that this sending is the first sending; Record the second sending time stamp of the data, and use the second sending time stamp and the system time stamp counter value as a benchmark reference point for all subsequent single-thread sending data. 6. the method for generating industrial ethernet packets based on the localization platform according to claim 5, wherein the file system is arranged in the Feiteng platform based on the localization, and the network card is a DMA network card based on the Feiteng platform . 7. A device for generating industrial Ethernet data packets based on a localization platform, characterized in that it comprises: The data file acquisition unit acquires several data files stored in the file system, each of which includes industrial Ethernet data to be simulated in a data frame type; The cache unit is used to store the data frame loads contained in the plurality of data files in a plurality of ring buffers in a one-to-one correspondence; the data creation thread is used to create and obtain a plurality of the ring buffers, and the number of the plurality of ring buffers is The same as the number of several data files, the number of nodes contained in each ring buffer is the same as the number of data contained in each data file, and the content of the nodes includes data load content, data load length, data transmission speed; The serialization recording unit is used to serialize the data frame loads in the ring buffer one by one by using the multi-core parallel serialization model to obtain a number of serialization records, and push the number of serialization records into the multi-producer multiple Consumer lock-free queue; the multi-core parallel serialization model includes several first parallel worker threads, the serialization records include the target serialization format, each of the first parallel worker threads is connected in series according to the ring buffer linked list Sequentially take out the data frame load in the ring buffer at a consumption speed that is multiple times that of the multi-producer multi-consumer lock-free queue and serialize to obtain serialized records; the multi-core parallel serialization model is single-threaded Working methods include: After determining the empty semaphore of the multi-producer multi-consumer lock-free queue, take out the load content, data load length, and data transmission speed related information from the corresponding ring buffer; Allocating memory from a memory pool to create the serialized record and filling the data length into the serialized record, the memory pool being created and obtained by the data creation thread; A sending unit, configured to take out several of the serialized records in the multi-producer and multi-consumer lock-free queue by using a parallel sending model, and send them outward through the network card in a direct memory access mode; the parallel sending model includes A number of second parallel worker threads.

Images