WO2024001340A1

WO2024001340A1 - Multicast packet duplication method and apparatus, storage medium, and electronic apparatus

Info

Publication number: WO2024001340A1
Application number: PCT/CN2023/083747
Authority: WO
Inventors: 王璞; 刘衡祁; 张自渊; 唐美芹
Original assignee: 深圳市中兴微电子技术有限公司
Priority date: 2022-06-30
Filing date: 2023-03-24
Publication date: 2024-01-04
Also published as: CN117376274A

Abstract

The present application discloses a multicast packet duplication method and apparatus, a storage medium, and an electronic apparatus. The method comprises: storing, into a multicast packet duplication queue, a multicast packet to be duplicated; determining, from the multicast packet duplication queue, a first multicast packet to be duplicated at which a target pointer points; searching for multicast member information of the first multicast packet, and acquiring first multicast duplication information of the first multicast packet; while waiting for receiving the multicast member information of the first multicast packet, enabling the target pointer to point at a next second multicast packet to be duplicated, so as to acquire multicast duplication information of the second multicast packet; and when all the multicast member information of the first multicast packet is received, moving the first multicast packet carrying the first multicast duplication information out of the multicast packet duplication queue.

Description

Multicast packet copying method and device, storage medium and electronic device

Related applications

This application claims priority to the Chinese patent application with application number 202210765523.X filed on June 30, 2022, the entire content of which is incorporated into this application by reference.

Technical field

Embodiments of the present application relate to the field of communications, and specifically, to a multicast packet copying method and device, a storage medium, and an electronic device.

Background technique

Currently, in the process of network communication, there is a need to send information from a source host to multiple other destination hosts in the network. In order not to waste bandwidth resources by sending information to hosts that do not need the information, it is necessary to use all possible means to avoid wasting bandwidth resources. Explicitly select some destination hosts in the destination to send, that is, multicast.

In the related technology, when multicast messages in a switching chip are queried for multicast members and copied, the first packet is copied first and then the next packet is copied. In the small packet scenario, due to the multicast table lookup delay, when the multicast table lookup delay is greater than the time it takes to copy the multicast packet once, full line-speed replication cannot be achieved, resulting in a waste of bandwidth.

Regarding related technologies, no effective solution has yet been proposed.

Contents of the invention

Embodiments of the present application provide a multicast packet copying method and device, a storage medium, and an electronic device to at least solve the problem in related technologies that the multicast table lookup delay is greater than the time it takes to copy the multicast packet once. Achieve full line-speed replication, thereby causing technical problems of bandwidth waste.

According to an embodiment of the present application, a method for copying a multicast packet is provided, which includes: storing the multicast packet to be copied in a multicast packet copy queue; and determining the target pointer from the multicast package copy queue. Point to the first multicast packet to be copied, where the above-mentioned multicast packet copy queue includes at least two multicast packets to be copied; search for the multicast member information of the above-mentioned first multicast packet, and obtain the above-mentioned first multicast packet The first multicast copy information of the packet; while waiting to receive the multicast member information of the above-mentioned first multicast packet, point the above-mentioned target pointer to the next second multicast packet to be copied to obtain the above-mentioned second multicast packet. multicast replication information; upon receiving all multicast member information of the first multicast packet, the first multicast packet carrying the first multicast replication information is removed from the multicast packet replication queue.

According to another embodiment of the present application, a device for copying multicast packets is provided, including: a storage module for storing multicast packets to be copied into a multicast packet copy queue; and a determination module for starting from The first multicast packet to be copied pointed to by the target pointer is determined in the above-mentioned multicast packet copy queue, wherein the above-mentioned multicast packet copy queue includes at least two multicast packets to be copied; the search module is used to find the above-mentioned third multicast packet. Multicast member information of the multicast packet, and obtain the first multicast copy information of the above-mentioned first multicast packet; the acquisition module is used to obtain the above-mentioned multicast member information while waiting to receive the multicast member information of the above-mentioned first multicast packet. The target pointer points to the next second multicast packet to be copied to obtain the multicast copy information of the above-mentioned second multicast packet; the removal module is used when all the multicast member information of the above-mentioned first multicast packet is received. Next, the first multicast packet carrying the first multicast replication information is moved out of the multicast packet replication queue.

According to yet another embodiment of the present application, a computer-readable storage medium is also provided, wherein the computer-readable storage medium A computer program is stored in the medium, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

According to yet another embodiment of the present application, an electronic device is also provided, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to perform any of the above. Steps in method embodiments.

Through this application, since the multicast packets to be copied can query multicast member information in turn, there is no need to wait for the completion of copying the previous packet. When each packet is copied out of the queue, it is judged whether the copy is completed. If the copy is not completed, it continues to return to the copy queue and takes turns querying the multicast member information with other multicast packets. Therefore, it is possible to solve the technical problem of being unable to replicate at full line speed due to the multicast table lookup delay being greater than the time it takes for the multicast packet to be copied once, thereby causing bandwidth waste, and achieve the technical effect of improving bandwidth utilization.

Description of drawings

Figure 1 is a schematic diagram of the system structure of an optional multicast packet copying method according to an embodiment of the present application;

Figure 2 is a schematic diagram of an optional multicast packet copying method according to an embodiment of the present application;

Figure 3 is a flow chart of an optional multicast packet copying method according to an embodiment of the present application;

Figure 4 is a schematic diagram of another optional multicast packet copying method according to an embodiment of the present application;

Figure 5 is a structural block diagram of an optional multicast packet copying device according to an embodiment of the present application.

Detailed ways

In order to enable those in the technical field to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only These are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts should fall within the scope of protection of this application.

It should be noted that the terms "first", "second", etc. in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances so that the embodiments of the application described herein can be practiced in sequences other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus that encompasses a series of steps or units and need not be limited to those explicitly listed. Those steps or elements may instead include other steps or elements not expressly listed or inherent to the process, method, product or apparatus.

In this embodiment of the present application, an embodiment of the above multicast packet copying method is provided. As an optional implementation manner, the multicast packet copying method can be applied to, but is not limited to, the system structure as shown in Figure 1 .

In one implementation, as shown in Figure 1, the system includes an "address generation device 101", a "message order preserving device 102", a "message inspection device 103", a "multicast copying device 104" and a "net charge storage device 105".

Address generation device 101: manages a set of linked lists of storage addresses, and the storage addresses correspond to the payload storage addresses in the payload storage device. When a multicast packet is input, the address generation device 101 allocates an address to the multicast packet and stores the multicast packet payload data in the corresponding address space in the payload storage device 105 . Before the multicast packet is output from the system, the multicast packet carrying only the necessary packet descriptor information such as the multicast packet storage address and multicast identifier is used instead of the multicast packet for processing in other devices.

Message sequence preserving device 102: used to restore the arrangement order between the internal cells of the multicast packet and the multicast packet. In this way, in the subsequent multicast replication device 104, the replication behavior of all cells in the same multicast packet can be guaranteed to be the same, and multicast packets can be avoided. Copy out of order.

Message checking device 103: used to check whether the message descriptor information carried in the multicast packet is true and correct. Multicast packets with errors in the check will be discarded directly. In this way, in the subsequent multicast duplication device 104, it can be ensured that the multicast packets participating in the duplication are correct multicast packets, and the duplication of erroneous multicast packets can be avoided to waste bandwidth ineffectively.

Payload storage device 105: used to store payload information of multicast packets. When a multicast packet carries multicast replication information and is ready to be output from the system, the multicast packet needs to extract the storage address allocated by the address generation device 101 and send it to the payload storage device, thereby extracting the multicast packet payload. Subsequently, the complete multicast packet can be output from the system carrying multicast replication information.

The multicast replication device 104 includes a multicast replication queue management device 203, a multicast replication table lookup device 202, a multicast replication modification device 201, and a multicast address recovery device 204. As shown in Figure 2, the multicast packet to be copied first enters the multicast copy queue management device 203 and is queued to wait for copying. At the same time, the multicast identifier is extracted from the first cell of the packet and sent to the multicast copy table lookup device 202 to query the multicast copy. information. After the multicast replication queue management device 203 receives the returned multicast replication information, the multicast packet can carry the multicast replication information and be output to the system where it is located, and participate in multicast replication port scheduling in the lower-level subsystem. The multicast packets that have completed multicast replication port scheduling are returned to the multicast replication address recovery device 204 along the original route, and then the multicast packets are sent to the payload storage device 105 . At the same time, the multicast address recycling device 204 will determine whether all the multicast packets have been copied. If all the copies have been completed, the payload storage device 105 will be informed that all cell addresses in the multicast packet can be recycled and can be used. Used by other multicast packets to enter this system.

This embodiment provides a method for copying multicast packets. Figure 3 is a flow chart of a method for copying multicast packets according to an embodiment of the present application. The process includes the following steps:

Step S302: Store the multicast packet to be copied in the multicast packet copy queue;

Step S304: Determine the first multicast packet to be copied pointed to by the target pointer from the multicast packet copy queue, where the multicast packet copy queue includes at least two multicast packets to be copied;

Step S306: Search for the multicast member information of the first multicast packet and obtain the first multicast replication information of the first multicast packet;

Step S308: While waiting to receive the multicast member information of the first multicast packet, point the target pointer to the next second multicast packet to be copied to obtain the multicast replication information of the second multicast packet;

Step S310: When all multicast member information of the first multicast packet is received, the first multicast packet carrying the first multicast replication information is removed from the multicast packet replication queue.

In one embodiment, a multicast replication queue is set up for the multicast packets to be copied. The multicast replication queue is composed of several registers. The number of registers should be enough to offset the table lookup delay of one multicast packet. .

In one embodiment, a table lookup pointer is set for the multicast replication queue. The table lookup pointer initially points to the head of the queue. When a multicast packet enters the queue, the table lookup pointer starts to extract the multicast packet first flag from the queue head. When there is no output from the multicast replication queue, the lookup pointer gradually moves backward from the head of the queue to extract the first flag of the multicast packet; when there is output from the multicast replication queue, because the data in the shift register continues to move forward, the position of the table lookup pointer does not change. changes, and continuously extracts the packet first flag from the same register. Multicast replication information may include but is not limited to outbound port number and global encapsulation index number.

In one implementation, after a fixed table lookup delay, the multicast packet at the head of the queue is the first to obtain multicast member information. The time interval for other multicast packets to obtain multicast member information is consistent with the time interval for initiating table lookup requests.

In one implementation, in addition to obtaining the multicast member information each time the table is looked up, the address of the next table lookup is also returned. If the address of the next table lookup is the same as the address of this table lookup, it indicates that the group returned this time is The broadcast member information is the last multicast member information.

In one implementation, when each multicast packet is dequeued, it needs to be determined whether all multicast member information has been found. When the multicast packet has found all multicast member information, the replication can be terminated after dequeuing with this copy information.

Through the solution provided by the embodiment of this application, the multicast packet to be copied is stored in the multicast packet copy queue; the first multicast packet to be copied pointed to by the target pointer is determined from the multicast packet copy queue, where, The multicast packet copy queue includes at least two multicast packets to be copied; searches for the multicast member information of the first multicast packet and obtains the first multicast copy information of the first multicast packet; while waiting to receive the first multicast packet While receiving the multicast member information of the packet, point the target pointer to the next second multicast packet to be copied to obtain the multicast copy information of the second multicast packet; after receiving all multicast members of the first multicast packet In the case of information, the first multicast packet carrying the first multicast replication information is removed from the multicast packet replication queue. Since multicast packets to be copied can query multicast member information in turn, there is no need to wait for the previous packet to be copied completely, which improves bandwidth utilization and solves the problem that the multicast table lookup delay is greater than the time it takes to copy the multicast package once. Time makes it impossible to achieve full line-speed replication, which in turn causes technical problems of wasted bandwidth.

In an exemplary embodiment, searching for the multicast member information of the first multicast packet includes: extracting the multicast identifier of the first multicast packet from the packet header identifier of the first multicast packet; sending a message based on the first multicast packet. A query request generated by the multicast identifier, where the query request is used to initiate a read request for multicast member information to the multicast replication information table; receive the found multicast member information of the first multicast packet, and search The multicast member information of the first multicast packet received is integrated into the message descriptor.

In one embodiment, because the cell slice copy behavior in the same multicast packet is exactly the same, when the table lookup pointer detects the multicast packet first flag, the multicast identifier can be extracted from the cell descriptor and copied to the multicast information. The table initiates a read request to query multicast member information. One read request queries the information of one multicast member. This multicast packet integrates the found multicast member information into the message descriptor and can be dequeued. The query request is only related to whether the packet first flag in the register pointed to by the lookup table pointer is valid. If the queue is blocked, the table lookup pointer will continue to move backward until it points to the last register at the end of the queue, and then stop.

Through the solution provided by the embodiment of this application, multicast member information is queried based on the multicast identifier, thereby improving the accuracy of the query.

In an exemplary embodiment, after searching for the multicast member information of the first multicast packet, the method further includes: when the currently found multicast member information of the first multicast packet is not the last multicast member information , move the target multicast packet carrying the first multicast copy information out of the multicast packet copy queue, and copy and write the packet body data of the target multicast packet into the target cache, where the multicast packets stored in the target cache will be Re-enter the multicast packet into the copy queue.

In one embodiment, a target cache fifo is defined. In order to ensure the continuity of the packet body of each multicast packet, the depth of the target cache fifo should be greater than the number of cells contained in the longest multicast packet. Because the next table lookup address obtained by this multicast packet is different from the current table lookup address, it means that this multicast packet has not found the last multicast member information. When the multicast packet is dequeued carrying this copy information, a copy of its packet body data needs to be written into the target cache fifo. For the first cell of the multicast packet written to the target cache fifo, the bit field where the multicast identifier is located needs to be replaced with the next table lookup address. Because the multicast packet content during the multicast replication process only contains address information stored in the payload storage device, the total size of the target cache fifo is very small.

Through the solution provided by the embodiment of the present application, when the multicast member information of the currently found first multicast packet is not the last multicast member information, the target multicast packet carrying the first multicast copy information is removed from the group. The multicast packet copy queue is copied and the packet body data of the target multicast packet is copied and written into the target cache, ensuring the accuracy of multicast copy.

In an exemplary embodiment, after copying and writing the packet body data of the target multicast packet into the target cache, the method further includes: if no new multicast packet to be copied is entered into the multicast packet copy queue, The multicast packets stored in the target cache are entered into the multicast packet copy queue. Before the target cache is cleared, the multicast packet copy queue will not enter new multicast packets to be copied.

In one implementation, if the queue does not input a new multicast packet to be copied from the previous module at this time, the target The cache fifo can read out the cached multicast packet data to the queue. During the reading process, the queue cannot input new multicast packets from the previous module until the target cache fifo is cleared. If the queue is inputting a new multicast packet to be copied at this time, the target cache fifo needs to wait for the end of the multicast packet to be input to the queue before initiating a read operation.

For example, as shown in Figure 4, step S401, the multicast packet enters the queue, and the table lookup pointer scans the first flag of the multicast packet one by one; step S402, extracts the multicast flag from the packet first cell and initiates a table lookup; step S403, copy After the information is returned, the corresponding multicast packet can be dequeued; step S404, determine whether the last copy of the multicast packet has been found; if so, jump to step S405; if not, jump to step S406; step S405, the group The copy of the broadcast packet ends; step S406, the multicast packet is written into the target cache; step S407, when no new multicast packet is entered into the queue, the multicast packet is read from the target cache and entered into the queue.

Through the solution provided by the embodiment of this application, when there are no new multicast packets to be copied entered into the multicast packet copy queue, the multicast packets stored in the target cache are entered into the multicast packet copy queue, which can ensure that the multicast packets are The continuity of the packet body can also prevent the total copying of a multicast packet from taking too long.

In an exemplary embodiment, obtaining the first multicast replication information of the first multicast packet includes: sending a table lookup request, where the table lookup request is used to request to obtain the multicast replication information based on the table lookup address; using the table lookup address Initiate a read request to the multicast replication information table; obtain one or more multicast replication information entries found in the multicast replication information table; parse the first group of one or more multicast replication information entries The broadcast replication information is packaged and sent to the multicast packet replication queue.

In one embodiment, after the multicast replication queue initiates a table lookup request, it sends the table lookup address to the information table for table lookup. In different multicast replication scenarios, the number of bits occupied by each copy of the information may be different. Limited by the RAM resources of the multicast replication information table, one copy of the information may need to correspond to multiple addresses, that is, multiple entries. In this case, it is necessary to divide the multicast replication information table RAM into multiple blocks according to depth, and configure multiple entries containing a copy of the replication information in different information table RAM blocks to facilitate that these entries can be read at the same time. out. The table lookup address corresponds to the RAM address of the information table where the first entry of each copied information is stored.

In one implementation, the table lookup address is used to initiate a read request to the multicast replication information table RAM (if a copy of the replication information is stored in multiple entries, a read request also needs to be initiated to multiple information table RAM blocks). After a fixed RAM read delay, one or more multicast replication information table entries are obtained.

In one embodiment, one or more multicast replication information table entries are sent to the parsing module for parsing to obtain multicast replication information such as outbound port numbers and global encapsulation index numbers.

In one implementation, the multicast replication information is packaged and returned to the multicast replication queue. Because the multicast replication queue may not be dequeued smoothly due to flow control by the downstream module, the multicast replication information needs to be temporarily stored in the parsing module. After waiting for the flow control on the multicast replication queue to be lifted, the parsing module sends the multicast replication information to the multicast replication queue one by one.

Through the solution provided by the embodiment of this application, one or more multicast replication information entries found in the multicast replication information table are obtained; the first multicast obtained by parsing one or more multicast replication information entries The replication information is packaged and sent to the multicast packet replication queue. Multicast replication information can be obtained through parsing.

In an exemplary embodiment, the method further includes: receiving a multicast change write request, wherein the multicast change write request is used to request to perform a change operation on the multicast replication information already stored in the multicast replication information table; temporarily storing the multicast Change the write request and start to monitor the multicast replication process of the multicast packet replication queue in real time; determine whether there is at least one multicast packet participating in the replication in the multicast packet replication queue; if there is at least one multicast packet participating in the replication, record When receiving a multicast change write request, the number of multicast packets that have not yet been sent for table lookup requests for querying multicast member information; When all packets are sent table lookup requests, table lookup requests are suspended, and the requested change table entry content is written to the corresponding write address according to the temporarily stored multicast change write request.

In one embodiment, the contents of the entries in the multicast replication information table are all delivered by the configuration bus. When the configuration bus has a need to add, delete or change the existing multicast replication information, it will send the multicast replication change device to the multicast replication information table. Deliver the multicast change write request, as well as the corresponding write address and write entry content.

In one embodiment, after receiving the multicast change write request, the multicast copy change device does not directly write the multicast change table entry content into the multicast copy table lookup device, but temporarily stores it first and starts multicast. Copy change preprocessing operations.

In one embodiment, after the multicast change entry content is temporarily stored, the multicast copy change device starts to monitor the multicast copy process in the multicast copy queue management device in real time, and determines whether there are still multicast packets in the queue participating in the copy.

In one embodiment, when there are still multicast packets in the queue participating in the replication, the number of multicast packets that have not yet initiated a table lookup in the multicast replication queue management device when the multicast change request is issued is recorded (that is, using The number of multicast packets in the multicast replication queue + the number of multicast packets in the target cache fifo - the number of multicast packets that have initiated a table lookup and are waiting for the replication information to be returned).

In one embodiment, when there are no multicast packets in the queue participating in the replication, it is determined whether the number of table lookup requests initiated by the multicast replication queue management device starting from the issuance of the multicast change request is equal to the recorded number of table lookup requests that have not yet been initiated. The number of multicast packets in the table. That is, it is determined whether table lookup has been initiated for all multicast packets.

In one embodiment, the paused multicast replication queue management device initiates a table lookup request, and sends the multicast change write request in the cache and the corresponding write address and write entry content to the multicast replication table lookup device. The playback table lookup device writes the content of the changed table entry into the information table.

In one embodiment, after the multicast copy and change device waits for all cached multicast change write requests to be written into the information table, it notifies the multicast copy queue management device of the end of the multicast change process, and the multicast copy queue restarts the table lookup and copy process. .

Through the solution provided by the embodiment of the present application, when no multicast packet participates in the replication, the sending of the table lookup request is suspended, and the requested change table entry content is written into the corresponding write request according to the temporarily stored multicast change write request. Enter the address. Ensure the accuracy of table lookup.

In an exemplary embodiment, when moving the first multicast packet carrying the first multicast replication information out of the multicast packet replication queue, the method further includes: recycling the multicast cell address in the first multicast packet; Determine whether the currently recycled multicast cell address is in the process of waiting for the read return of the multicast cell address with the same total number of copies; in the process of waiting for the read return of the multicast cell address with the same total number of copies, cancel the current Output the multicast cell corresponding to the recycled multicast cell address, and wait for the total number of copies to be read; when it is not in the process of waiting for the read return of the multicast cell address with the same total number of copies, use the currently recycled Use the multicast cell address as the read address to initiate a read request to the total number of copies; use the currently recycled multicast cell address as the write address, initiate a write request to the write address for the total number of copies.

In one embodiment, when the multicast cell carrying multicast replication information is output from the multicast queue management module to the lower-level system for port scheduling, the multicast cell address will be sent to the multicast address recycling device, where the port Scheduling is to determine the order in which multicast packets carrying different port numbers are sent to multiple destination hosts based on the priority of the port numbers.

In one embodiment, two total copy numbers RAM0 and RAM1 of the same size are defined to record the number of copies of each cell address. The write enable, write address and write data of RAM0 and RAM1 are consistent. Determine whether it is in the process of waiting for the read return of the same multicast cell address as the total copy number RAM0.

In one embodiment, while waiting for the read return of the same multicast cell address as the total copy number RAM0, the total copy number RAM0 read request is not initiated for the output of this multicast cell, and the register Record this cancellation behavior and wait for the read return of the same multicast cell address of RAM0, the total number of copies.

In one embodiment, when the read return process of the same multicast cell address as the total copy number RAM0 is not in progress, a read request is initiated to the total copy number RAM0 using the multicast cell address as the read address.

In one embodiment, the read data of the total copy number RAM0 is the number of copies of the multicast cell.

In one implementation, the address of the multicast cell is used as the write address, and (the number of copies of the multicast cell + the number of canceled read requests during the waiting for read return process + 1) is used as the write data, and the address is written to the total address. Copies RAM0 and RAM1 initiate write requests.

Through the solution provided by the embodiment of this application, the currently recycled multicast cell address is used as the write address, and a total copy number write request is initiated to the write address. You can record the number of multicast packets output from the multicast queue management module to the lower-level system for port scheduling.

In an exemplary embodiment, when recycling the multicast cell address in the first multicast packet, it also includes: determining whether the currently recycled multicast cell address is waiting for the same number of multicast cells that have been scheduled. During the read return process of the address; while waiting for the read return of the multicast cell address with the same number of scheduled copies, cancel the output of the multicast cell corresponding to the currently recycled multicast cell address, and initiate the scheduled copy number and the total number of copies; when the read return process is not waiting for the same multicast cell address of the scheduled copies, the currently recycled multicast cell address is used as the read address, and at the same time, the read request to the scheduled copies is and the total number of replicas to initiate a read request.

In one embodiment, after the lower-level system completes multicast replication outbound port scheduling, it returns the multicast cell content to the multicast replication module and sends it to the multicast address recycling device.

In one embodiment, a scheduled copy RAM with the same size as the total copy number RAM1 is defined to record the number of times each cell address has participated in scheduling. The read enable and read address of the scheduled copy RAM and the total copied copy RAM1 are consistent. Determine whether it is in the process of waiting for the read return of the same multicast cell address of the scheduled RAM.

In one embodiment, while waiting for the read return of the same multicast cell address of the scheduled copy RAM, the scheduled copy RAM and total copy number are not initiated for the output of this multicast cell. RAM1 reads the request, and uses a register to record the cancellation behavior, waiting for the read return of the same multicast cell address of the scheduled copies RAM and the total number of copies RAM1.

In one embodiment, when the read return process of the same multicast cell address of the scheduled copy RAM is not in the process, the multicast cell address is used as the read address to simultaneously copy the scheduled copy RAM and the total copy address. Copy RAM1 initiates a read request.

In one embodiment, the read data of the scheduled copies RAM is the number of copies of the multicast cell, and the read data of the total copy RAM1 is the number of times the multicast cell has been scheduled.

In one implementation, after writing once, it is determined whether the number of copied copies is equal to (the number of times scheduling has been completed + 1).

In one embodiment, when the number of copied copies is equal to (the number of times scheduling has been completed + 1), the multicast cell content is sent back to the payload storage device, and an address recycling flag is given to inform the payload. The cell corresponding to the address of the storage device has completed all copy operations and can be released. At the same time, the multicast cell address is used as the write address, and 0 is the write data to initiate a write request to the scheduled copy RAM to avoid affecting the recycling of the cell address in the payload storage device.

In one implementation, when the number of copied copies is not equal to (number of times of completed scheduling + 1), the multicast cell content is sent back to the payload storage device. At the same time, the address of the multicast cell is used as the write address, and (the number of times the multicast cell has been scheduled + the number of read requests canceled while waiting for read return + 1) is used as the write data, and the number of RAMs that have been scheduled is written. Initiate a write request.

Through the solution provided by the embodiments of this application, by initiating read requests to the number of scheduled copies and the total number of copies, it can be judged whether the total number of scheduled copies is equal to the total number of copies, that is, it can be judged whether the number of multicast packets returned after completing the scheduling is Whether the number of copies is equal to the number of multicast packets for port scheduling.

Through the description of the above embodiments, those skilled in the art can clearly understand the methods according to the above embodiments. The method can be implemented with the help of software plus the necessary general hardware platform. Of course, it can also be implemented through hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence or that contributes to the existing technology. The computer software product is stored in a storage medium (such as ROM/RAM, disk, CD), including several instructions to cause a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of this application.

This embodiment also provides a device for copying multicast packets, which is used to implement the above embodiments and preferred implementations. What has already been described will not be described again. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

Figure 5 is a structural block diagram of a device for replicating multicast packets according to an embodiment of the present application. As shown in Figure 5, the device includes:

The storage module 501 is used to store multicast packets to be copied into the multicast packet copy queue;

The determination module 502 is configured to determine the first multicast packet to be copied pointed by the target pointer from the multicast packet copy queue, where the multicast packet copy queue includes at least two multicast packets to be copied;

The search module 503 is used to search for the multicast member information of the first multicast packet and obtain the first multicast replication information of the first multicast packet;

The acquisition module 504 is configured to point the target pointer to the next second multicast packet to be copied while waiting to receive the multicast member information of the first multicast packet, so as to obtain the multicast replication information of the second multicast packet;

The removal module 505 is configured to remove the first multicast packet carrying the first multicast replication information from the multicast packet replication queue when all multicast member information of the first multicast packet is received.

It should be noted that each of the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but is not limited to this: the above modules are all located in the same processor; or the above modules can be implemented in any combination. The forms are located in different processors.

Embodiments of the present application also provide a computer-readable storage medium that stores a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

In an exemplary embodiment, the computer-readable storage medium may include but is not limited to: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM) , mobile hard disk, magnetic disk or optical disk and other media that can store computer programs.

An embodiment of the present application also provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

In an exemplary embodiment, the above-mentioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the above-mentioned processor, and the input-output device is connected to the above-mentioned processor.

For specific examples in this embodiment, reference may be made to the examples described in the above-mentioned embodiments and exemplary implementations, and details will not be described again in this embodiment.

Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present application can be implemented using general-purpose computing devices, and they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices. They may be implemented in program code executable by a computing device, such that they may be stored in a storage device for execution by the computing device, and in some cases may be executed in a sequence different from that shown herein. Or the described steps can be implemented by making them into individual integrated circuit modules respectively, or by making multiple modules or steps among them into a single integrated circuit module. As such, the application is not limited to any specific combination of hardware and software.

The above are only optional embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the principles of this application shall be included in the protection scope of this application.

Claims

A method for copying multicast packets, including:

Store the multicast packets to be copied in the multicast packet copy queue;

Determine the first multicast packet to be copied pointed to by the target pointer from the multicast packet copy queue, where the multicast packet copy queue includes at least two multicast packets to be copied;

Search for the multicast member information of the first multicast packet, and obtain the first multicast replication information of the first multicast packet;

While waiting to receive the multicast member information of the first multicast packet, point the target pointer to the next second multicast packet to be copied to obtain the multicast replication information of the second multicast packet;

When all multicast member information of the first multicast packet is received, the first multicast packet carrying the first multicast replication information is removed from the multicast packet replication queue.
The method according to claim 1, wherein the finding the multicast member information of the first multicast packet includes:

Extract the multicast identifier of the first multicast packet from the packet header identifier of the first multicast packet;

Send a query request generated based on the multicast identifier of the first multicast packet, wherein the query request is used to initiate a read request query for multicast member information to the multicast replication information table;

Receive the found multicast member information of the first multicast packet, and integrate the found multicast member information of the first multicast packet into a message descriptor.
The method according to claim 1, wherein after searching for the multicast member information of the first multicast packet, it further includes:

When the currently found multicast member information of the first multicast packet is not the last multicast member information, the target multicast packet carrying the first multicast copy information is moved out of the multicast The packet copy queue is configured to copy and write the packet body data of the target multicast packet into the target cache, where the multicast packets stored in the target cache will be re-entered into the multicast packet copy queue.
The method according to claim 3, wherein after copying and writing the packet body data of the target multicast packet into the target cache, it further includes:

In the case where no new multicast packet to be copied is entered into the multicast packet copy queue, the multicast packet stored in the target cache is entered into the multicast packet copy queue, wherein, when the target cache is Before clearing, the multicast packet copy queue will not input new multicast packets to be copied.
The method according to claim 2, wherein said obtaining the first multicast replication information of the first multicast packet includes:

Send a table lookup request, where the table lookup request is used to request to obtain multicast replication information based on the table lookup address;

Use the table lookup address to initiate a read request to the multicast replication information table;

Obtain one or more multicast replication information entries found in the multicast replication information table;

The first multicast replication information obtained by parsing the one or more multicast replication information entries is packaged and sent to all The multicast packet replication queue.
The method of claim 5, further comprising:

Receive a multicast change write request, wherein the multicast change write request is used to request to perform a change operation on the multicast replication information already stored in the multicast replication information table;

Temporarily store the multicast change write request, and start monitoring the multicast replication process of the multicast packet replication queue in real time;

Determine whether there is at least one multicast packet participating in the replication in the multicast packet replication queue;

In the case where at least one multicast packet participates in replication, record the number of multicast packets that have not yet been sent for table lookup requests for querying multicast member information when the multicast change write request is received;

When all the multicast packets of the query request for querying multicast member information that have not yet been sent are sent to the table lookup request, suspend sending the table lookup request, and write the multicast change request according to the temporary storage, Write the requested change entry content to the corresponding write address.
The method according to claim 1, wherein when moving the first multicast packet carrying the first multicast replication information out of the multicast packet replication queue, it further includes:

Recycle the multicast cell address in the first multicast packet;

Determine whether the currently recycled multicast cell address is in the process of waiting for the read return of the multicast cell address with the same total number of copies;

While waiting for the read return of the multicast cell address with the same total number of copies, cancel the output of the multicast cell corresponding to the currently recycled multicast cell address, and wait for the total number of copies. read request;

While not in the process of waiting for the read return of the multicast cell address with the same total number of copies, use the currently recycled multicast cell address as the read address to initiate a read request to the total number of copies;

Using the currently recycled multicast cell address as a write address, initiate a write request for the total number of copies to the write address.
The method according to claim 7, wherein when recycling the multicast cell address in the first multicast packet, it further includes:

Determine whether the currently recycled multicast cell address is in the process of waiting for the read return of the same number of scheduled multicast cell addresses;

While waiting for the read return of the multicast cell address with the same number of scheduled copies, cancel the output of the multicast cell corresponding to the currently recycled multicast cell address, and initiate the number of scheduled copies and total copies. number of read requests;

While not waiting for the read return of the multicast cell address with the same number of scheduled copies, use the currently recycled multicast cell address as the read address, and simultaneously initiate a read request to the scheduled number of copies and the total number of copies. .
A device for replicating multicast packets, including:

The storage module is set to store multicast packets to be copied into the multicast packet copy queue;

Determining module, configured to determine the first multicast to be copied pointed to by the target pointer from the multicast packet copy queue. packet, wherein the multicast packet copy queue includes at least two multicast packets to be copied;

A search module configured to search for the multicast member information of the first multicast packet and obtain the first multicast replication information of the first multicast packet;

The acquisition module is configured to point the target pointer to the next second multicast packet to be copied while waiting to receive the multicast member information of the first multicast packet, so as to obtain the information of the second multicast packet. Multicast replication information;

A removal module configured to remove the first multicast packet carrying the first multicast copy information out of the multicast packet copy when all multicast member information of the first multicast packet is received. queue.
A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute the method described in any one of claims 1 to 8 when running. .
An electronic device including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to perform the method described in any one of claims 1 to 8. method.