A cross-switch chip ACL management method and system

By supporting kernel-state and user-state configuration ACL rules, priority attributes and instance management mechanisms, cross-platform ACL resource management is realized, solving the flexibility and scalability of ACL management of different switching chips, and improving the flexibility of communication systems and development efficiency.

CN114416395BActive Publication Date: 2024-12-06TRANSCEND COMM BEIJING
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111614599.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-27
Publication Date
2024-12-06
Estimated Expiration
2041-12-27

AI Technical Summary

Technical Problem

The prior art is difficult to implement cross-platform management of ACLs of different switching chips, and the priority configuration and instance management mechanism of ACL rules are not flexible enough, which affects the flexibility and scalability of the communication system.

Method used

Provides a cross-switch chip ACL management method, supports kernel-state and user-state configuration ACL rules, supports ACL priority attributes and instance management mechanisms, and realizes ACL configuration and instance management through a unified communication interface, including the transfer of ACL configuration information in the pseudo-file system. , priority search and movement and black hole recovery mechanisms when inserting and deleting configurations to ensure efficient management of ACL resources.

Benefits of technology

It realizes cross-platform unified management of ACL resources, improves the flexibility and scalability of the communication system, supports developers to configure ACL priorities and instances on demand, reduces development costs, and simplifies system maintenance and expansion.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114416395B_ABST
    Figure CN114416395B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of communication technology, and in particular to a cross-switch chip ACL management method and system thereof. The method includes supporting kernel state and user state configuration of ACL rules, supporting ACL priority attributes, supporting ACL instance management mechanism, and supporting a unified communication interface. The present invention provides an ACL priority configuration method, a communication method for sending ACL configuration from user state to user state and from kernel state to user state, and enables communication system software to manage ACL resources across platforms and switch chips through the ACL instance management mechanism, and provides a unified programming interface.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a cross-switch chip ACL management method and system thereof. Background Art

[0002] The order of ACL (Access Control List) statements determines the order of data control. ACL statements are arranged logically from top to bottom. During the data matching process, statements are compared one by one. Once a match is successful, it is processed according to the current statement control policy and is no longer compared with subsequent statements. Therefore, the correct statement order can achieve the desired control effect, and increasing the priority configuration of rule statements can improve the flexibility of using ACL.

[0003] Different switching chips implement ACL functions in different ways. An adaptation layer is added between the user interface and the switching chip interface to shield the details of different switching chips and implement cross-platform management of switching chip ACL. This is also the location of the implementation of this application solution in the communication system software framework. The longest matching field length of ACL rules is different. For some switching chips, one rule can meet the configuration requirements of all command lines, while some switching chips require multiple ACL rules to be assembled to meet the requirements of one command line. It is necessary to make the process of calling the switching chip ACL interface transparent to achieve a cross-platform effect; the priorities of different types of ACL rules are different, and the priority of protocol ACL rules is lower than that of command line ACL rules.

[0004] This application belongs to the middle layer of the communication system adaptation switching chip API (Application Programming Interface), belongs to the user state, and needs to obtain the data sent by the command line from the kernel state of the communication system.

[0005] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a cross-switch chip ACL management method and system thereof to solve the technical problems existing in the prior art.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] The present invention provides a cross-switch chip ACL management method, which includes supporting kernel state and user state configuration of ACL rules, supporting ACL priority attributes, supporting ACL instance management mechanism, and supporting a unified communication interface;

[0009] Among them, support kernel mode and user mode configuration ACL rules, including:

[0010] In the communication system, the user-mode module sends the ACL configuration information through the socket, and the kernel-mode module transmits the ACL configuration information through the pseudo file system;

[0011] The supported ACL priority attributes include:

[0012] Users configure priorities based on project needs to determine the order in which ACLs take effect;

[0013] Among them, the ACL instance management mechanism is supported, including:

[0014] When inserting a configuration, find the appropriate ACL instance number to be issued according to the configuration priority, move the configuration before or after the insertion point, and insert the new configuration; when deleting a configuration, find the ACL instance number used by the configuration to be deleted, and set the bitmap data structure after deletion; the black hole recovery mechanism includes black hole recovery during insertion and timed black hole recovery. When recovering a black hole during insertion, try to reduce the number of black holes as much as possible, and timed black hole recovery ensures that the black hole occupancy rate remains below the threshold, which can converge the number of black holes;

[0015] Among them, unified communication interfaces are supported, including:

[0016] The pack / unpack interface for encapsulating / parsing messages, the msgUserSend / msgUserRecv interface for the communication library to send / receive messages from user mode, the msgKernelSend interface for the communication library to send messages from the kernel, and the communication library to obtain the length values ​​of all ACL field pairs supported by the switching chip from user mode / kernel mode using getTypeLengthFromUser / getTypeLengthFromKernel.

[0017] Preferably, the daemon process adds the configuration message sent by the communication library to the end of the message processing queue, and returns the processing result after the message is processed, waking up the waiting caller, thereby realizing synchronous communication.

[0018] Preferably, when the communication library is called for the first time, all ACL field length values ​​are obtained at one time.

[0019] Preferably, the method for inserting an ACL configuration into an instance block includes:

[0020] First, according to the ACL configuration priority, find the location i to be inserted;

[0021] Secondly, find the first free instance positions b and a before and after the insertion point respectively;

[0022] Again, compare ib with ai. If ib is smaller, move ib instances before the insertion point i forward in sequence; otherwise, move ai instances after the insertion point i backward in sequence.

[0023] Finally, the configuration is inserted into the position i to be inserted.

[0024] Preferably, the method for reclaiming ACL instance resources includes: reclaiming the black hole during insertion and reclaiming the black hole at a scheduled time. Reclaiming the black hole during insertion is a real-time best-effort solution, and reclaiming the black hole at a scheduled time is an asynchronous solution to ensure black hole convergence. The two solutions are used together to maintain the black hole within an acceptable range.

[0025] Recycling black holes during insertion is to reduce the existing black holes as much as possible when inserting the ACL configuration. Its mechanism adopts the method of inserting the ACL configuration into the instance block.

[0026] Timed black hole recovery: When the black hole occupancy ratio exceeds the preset threshold, the timer is started. After the timer times out, if the black hole occupancy ratio still exceeds the preset threshold, the recovery process is started. First, the location of the black hole is obtained in order. Second, the black hole points are filled in order from small to large. Finally, the above two processes are executed in the entire ACL block partition.

[0027] Preferably, the developer divides the instance blocks as needed; when an ACL rule has too many configuration attributes, the instance aggregation function of the present invention is enabled, and the basic unit of instance management is changed from a physical instance to a logical instance composed of two physical instances.

[0028] The present invention provides a cross-switch chip ACL management system, which adopts the cross-switch chip ACL management method.

[0029] By adopting the above technical solution, the present invention has the following beneficial effects:

[0030] The present invention provides an ACL priority configuration method, a communication method for sending ACL configuration from user state to user state and from kernel state to user state, enables communication system software to manage ACL resources across platforms and switching chips through an ACL instance management mechanism, and provides a unified programming interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A system framework diagram provided for an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of the TLV organization form of communication data provided by an embodiment of the present invention;

[0034] Figure 3 A schematic diagram of a socket communication C / S model provided in an embodiment of the present invention;

[0035] Figure 4 A schematic diagram of a pseudo file communication model provided by an embodiment of the present invention;

[0036] Figure 5 A schematic diagram of an example block division diagram provided in an embodiment of the present invention;

[0037] Figure 6 A schematic diagram of an insertion mechanism provided by an embodiment of the present invention;

[0038] Figure 7 A flow chart of improved configuration of movement during insertion provided by an embodiment of the present invention;

[0039] Figure 8 A timed black hole recovery diagram provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0042] Embodiment 1

[0043] Combination Figures 1 to 8As shown, this embodiment provides a cross-switch chip ACL management method, which includes supporting kernel state and user state configuration of ACL rules, supporting ACL priority attributes, supporting ACL instance management mechanism, and supporting a unified communication interface; wherein, supporting kernel state and user state configuration of ACL rules includes: in a communication system, the user state module sends ACL configuration information through a socket, and the kernel state module transmits ACL configuration information through a pseudo file system; wherein, supporting ACL priority attributes includes: the user configures the priority according to the project needs to determine the effective order of the ACL; wherein, supporting the ACL instance management mechanism includes: when inserting a configuration, searching for a suitable ACL instance number to be issued according to the configuration priority, moving the configuration before or after the insertion point, and inserting a new configuration; when deleting a configuration, searching for the configuration to be deleted for the user ACL instance number, set the bitmap data structure after deletion; black hole recovery mechanism, including black hole recovery during insertion and timed black hole recovery. Black hole recovery during insertion tries its best to reduce the number of black holes, and timed black hole recovery ensures that the black hole occupancy rate remains below the threshold and can converge the number of black holes; among them, support unified communication interfaces, including: pack / unpack interface for encapsulating / parsing messages, msgUserSend / msgUserRecv interface for the communication library to send / receive messages from user state, msgKernelSend interface for the communication library to send messages from the kernel, and the communication library obtains the length value getTypeLengthFromUser / getTypeLengthFromKernel for all ACL field pairs supported by the switching chip from user state / kernel state.

[0044] Embodiment 2

[0045] This embodiment provides a cross-switch chip ACL management system, which adopts the cross-switch chip ACL management method. This embodiment divides the cross-switch chip ACL management implementation system into six major functional modules, including a useful package (utils), a communication layer (communication), a communication library (communication lib), a mapping layer (mapping), an instance management layer (management) and a call layer (call), such as Figure 1As shown. The utils utility package includes basic data structures (list, queue, hash) and their basic operations (add, delete, modify, and query), data TLV encapsulation and parsing, etc. The communication layer receives data from other modules of the communication system, acts as a server in the user-mode socket mode, continuously monitors the links of other modules, and continuously and synchronously reads ACL configuration data in the kernel-mode data transmission mode. The communication library serves as the entry point for other functional modules of the communication system to communicate with the daemon process, completes the distribution of ACL configuration, and corresponds to the communication layer. The mapping layer completes the one-to-one mapping between ACL configuration data and ACL instances, and can quickly locate the actual effective ACL instance through the configuration data. The instance management layer rearranges the order of ACL instances according to the priority attribute, with the highest priority being placed in front, to realize the ACL priority function. The call layer mainly adapts the daemon process and the switching chip API data structure, shields the switching chip data structure details, and completes the daemon process initialization work.

[0046] After comprehensive analysis of the implementation of ACL by existing communication equipment manufacturers, the present invention integrates the classification of ACL by the above communication manufacturers as much as possible, and divides ACL into reserved ACL, layer 2 ACL, layer 3 ACL, hybrid ACL, custom template ACL, protocol ACL and reserved ACL. If the above classification is further classified by priority, protocol ACL is a category by itself, and the others are a category, and the protocol ACL has a low priority. At the same time, ACL is also divided into outbound and inbound directions in terms of use.

[0047] First, the encapsulation and parsing of communication data in the useful package are described in detail, while the basic data structure and its use in the useful package are not described.

[0048] The organization of communication data is based on the TLV encapsulation concept, and its implementation is as follows: Figure 2 As shown. In the first-layer TLV structure (referred to as Class I ACL), the above ACL classification type is used as "T". Under each Class I ACL classification, the type of the ACL setting field is used as "T" in the second-layer TLV structure (referred to as Class II ACL); there can be multiple Class I TLVs in a message, and each Class I TLV can contain multiple Class II ACLs. Class III TLV, Class IV TLV, etc. can be expanded later according to the situation. Note that the first Class I TLV of each message is fixed to fill in the ACL action type, and one message is an ACL rule. Provide a unified programming interface pack() / unpack() to the outside.

[0049] It is worth noting that different ACL field types correspond to different lengths, that is, the length of the second-class ACL varies according to the type. Therefore, before the communication library starts sending ACL configuration messages, it is necessary to obtain all the length values ​​corresponding to the ACL field types supported by the chip from the communication layer and store them for use when sending ACL configuration messages later.

[0050] Secondly, the implementation of the communication layer and communication library is described in detail. According to the different communication forms, they are divided into socket communication and pseudo-file communication.

[0051] Socket communication is based on the C / S model to complete inter-process communication. The daemon starts the socket listening. When other functional module A in the communication system calls the shared library to configure ACL, the daemon establishes a link with it and receives the ACL configuration message. The daemon puts the message at the end of the message queue. After the processing thread processes the message, it returns the processing result to module A. The processing flow is as follows: Figure 3 shown.

[0052] Pseudo-file communication is used to obtain ACL configuration messages from kernel-mode modules. Since kernel-mode resources are common to the entire system, attention must be paid to resource (configuration message) synchronization issues. The communication library maintains a message queue in kernel mode, inserts messages to be sent to the daemon process into the tail of the kernel message queue, and waits for the user mode to obtain, process, and return one by one. The daemon process creates a thread, process A, that reads the contents of the file (which must be created by itself) in the / proc directory through ioctl, obtains the message at the head of the kernel's message queue, copies the data to the user mode, adds it to the tail of the daemon process's message queue, and sleeps process A at the same time. After the daemon process processes the message queue, it wakes up process A, passes the processing results to the kernel again through ioctl, wakes up the kernel task, reads the head of the kernel message queue, and processes the next ACL configuration message. The processing flow is as follows: Figure 4 shown.

[0053] The communication between the communication library and the communication layer is synchronous. The communication layer maintains the message queue to avoid data asynchrony caused by multiple processes competing for communication. The communication between the communication library and the communication layer is comprehensive, supporting not only communication between user-state processes, but also communication between user-state and kernel-state.

[0054] Finally, since the mapping layer, instance management layer, and call layer are closely related, they are described together in detail.

[0055] When the communication system is initialized, the ACL function of the switching chip needs to be initialized, and the classification of the block where the ACL instance is located is particularly important. In order to shield the implementation details of the ACL instance block of different switching chips, the present invention proposes two sets of block division schemes. The implementation is mainly in the call layer.

[0056] In the first solution, all instance blocks do not distinguish between ACL uses. ACLs of different uses can be set in the entire block. The order of ACL instance arrangement depends on the priority, which is used as the basis for different ACL divisions. The higher the priority, the higher it is in the block; if the priorities are the same, the ACL instance issued first is in the front of the block. The advantage of this solution is that it is simple to implement, but the disadvantage is that it has high requirements for user usage priority.

[0057] The second solution is to first divide the instance blocks according to different ACL uses, and then add priority configuration options based on the divided blocks. That is, the priority is a local variable of the block, such as Figure 5 The advantage of this solution is that users can use it easily without having to worry about global priority division; the disadvantage is that it is complex to implement and has poor subsequent scalability.

[0058] The mapping layer is the link between the external ACL configuration data and the driver data issued by the present invention, maintaining the one-to-one correspondence between them, and facilitating the user to perform basic operations of adding, deleting, modifying and checking. The reason why it is used as a separate functional layer of the present invention is mainly to shield the details of the ACL configuration data and the differences in ACL commands. Its main function is to parse the ACL configuration message and manage the parsed data in the form of a hash-table. For data at the level of thousands, the hash-table data structure has good performance in data addition, deletion, modification and query.

[0059] The instance management layer is the key point of the present invention. It is managed based on the divided instance blocks. The bitmap data structure is used to record the instance usage. It is the core of maintaining and managing ACL and shielding the details of the ACL function of the switching chip. The basic idea of ​​management is that all ACL instance configurations to be issued to the switching chip are strictly set in the order of the higher the priority, the higher the priority; if the priority is the same, the ACL instance is issued first and set in front of the block, and the later one is issued and set at the back.

[0060] Some switching chips support the aggregation of two connected instances into a large logical instance. The new logical instance can match more ACL fields, thereby achieving more accurate ACL matching rules. However, some switching chips do not support the instance aggregation function. In order to adapt to all chips, the present invention supports developers to configure this attribute according to the chip function. If instance aggregation is not configured, the basic unit of instance management is a single physical instance; if instance aggregation is configured, the basic unit of instance management is a logical instance aggregated from two physical instances.

[0061] Assume that the switch chip has N blocks and each block has M instances. The following takes n consecutive blocks as objects and describes the instance management mechanisms such as insertion, deletion, and black hole recovery in detail without configuring instance aggregation.

[0062] Now it is required to configure four configurations c1, c2, c3 and c4 with priorities of 20, 30, 10 and 20 respectively. The operation process is as follows Figure 6 As shown. c1 is inserted first and is directly placed in the position of instance 1, as shown Figure 6 As shown in (A) in the figure; when c2 is inserted, since the priority of c2 is greater than the priority of c1, c1 and its subsequent configurations are moved back one position, and c2 is inserted before c1, as shown in the figure Figure 6 As shown in (B) in the figure; when inserting c3, ​​since c3 is less than all configured priorities, c3 can be directly inserted after c1, as shown in Figure 6 As shown in (C) in ; when c4 is inserted, its priority is greater than c3 and equal to c1, so c4 should be placed before c3 and after c1, as shown in Figure 6 As shown in (D) in the figure. Assuming that a configuration c5 with a priority of 50 is configured, its insertion principle is the same as above. First, the priorities are compared and it is found that its priority is greater than that of the configuration c2 in the first instance. Therefore, the first four configurations are moved back one position in turn and c5 is inserted at the first instance. The final configuration is that from the first instance to the fifth instance, the arrangement order is c5, c2, c1, c4, c3, strictly sorted by priority.

[0063] In the process of insertion, from the perspective of efficiency, the smaller the number of mobile configurations, the better. Therefore, a small optimization is made in the above insertion principle to minimize the number of mobile configurations. That is, when the position of the configuration to be inserted is found, compare the number of configurations before the position to be inserted, number_before, and the number of configurations after the position to be inserted, number_after. If number_before is greater than number_after, and there are enough free spaces behind for movement, move backward; if number_before is less than number_after, and there are enough free spaces in front for movement, move forward; if number_before is equal to number_after, you can move forward or backward. More details are as follows Figure 7 shown.

[0064] The deletion action is relatively simple. Through the ACL configuration to be deleted, find the instance number where the configuration actually takes effect, directly call the instance deletion function, and clear the bitmap data corresponding to the configuration. In this way, the general deletion process is basically completed.

[0065] After deletion, unused black holes will be left in the bitmap structure. These black holes need to be recycled. After multiple insertions and deletions, the scarce ACL instance resources may leave large black holes, wasting resources. The present invention proposes two solutions, namely, recycling black holes during insertion and recycling black holes at a fixed time.

[0066] Reclaim black holes during insertion, that is, in the insertion process, after determining the position i to be inserted, when starting to perform the move operation, it is necessary to find the black hole point b closest to the point to be inserted. If moving backward, take the black hole point b as the tail of the movement and the point to be inserted i as the head, and start moving backward; if moving forward, take the black hole point b as the head of the movement and the point to be inserted i as the tail, and start moving forward. The advantage of this method is that it can reduce the number of moves and the number of black holes. The disadvantage is that when the insertion process does not involve moving configuration, the number of black holes will not be less.

[0067] Therefore, a method of regularly recycling black holes is needed to ensure that the number of black holes remains within a reasonable range. The specific mechanism is to set a black hole recycling threshold for the ratio of black holes to the total number of management block instances. When the threshold is exceeded, the timer is started. If the ratio is still greater than the threshold after the timer times out, the recycling process is started. That is, based on the recorded black hole points, the black holes are reduced from small to large. The specific process is as follows Figure 8 It is worth noting that the black hole recovery during insertion and the scheduled black hole recovery need to be used at the same time. The black hole recovery during insertion is to recover black holes as much as possible, while the scheduled black hole recovery ensures that the number of black holes in the ACL instance resources always converges to a reasonable value.

[0068] In summary, through the present invention, an intermediate layer for adapting the switching chip ACL interface call is added to the communication system, supporting comprehensive communication modes in user mode and kernel mode, supporting developers to customize ACL blocks for on-demand classification, supporting priority configuration attributes, and providing developers with a unified programming interface, and finally realizing the cross-switching chip and cross-command line platform characteristics of the present invention, making large-scale communication software systems easy to develop, maintain and expand, saving development costs.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A cross-switch chip ACL management method, characterized in that: Including support for kernel-mode and user-mode configuration of ACL rules, support for ACL priority attributes, support for ACL instance management mechanism, and support for a unified communication interface; Among them, support kernel mode and user mode configuration ACL rules, including: In the communication system, the user-mode module sends the ACL configuration information through the socket, and the kernel-mode module transmits the ACL configuration information through the pseudo file system; The supported ACL priority attributes include: Users configure priorities based on project needs to determine the order in which ACLs take effect; Among them, the ACL instance management mechanism is supported, including: When inserting a configuration, find the appropriate ACL instance number to be issued according to the configuration priority, move the configuration before or after the insertion point, and insert the new configuration; when deleting a configuration, find the ACL instance number used for the configuration to be deleted, and set the bitmap data structure after deletion; the black hole recovery mechanism includes black hole recovery during insertion and timed black hole recovery. Black hole recovery during insertion reduces the number of black holes, and timed black hole recovery ensures that the black hole occupancy rate remains below the threshold, which can converge the number of black holes; Among them, unified communication interfaces are supported, including: The pack / unpack interface for encapsulating / parsing messages, the msgUserSend / msgUserRecv interface for the communication library to send / receive messages from user state, the msgKernelSend interface for the communication library to send messages from the kernel, and the communication library to obtain the length value getTypeLengthFromUser / getTypeLengthFromKernel for all ACL fields supported by the switching chip from user state / kernel state.

2. The cross-switch chip ACL management method according to claim 1, characterized in that: The daemon process adds the configuration message sent by the communication library to the end of the message processing queue. After the message is processed, it returns the processing result and wakes up the waiting caller, thus realizing synchronous communication.

3. The cross-switch chip ACL management method according to claim 1, characterized in that: When the communication library is called for the first time, all ACL field length values ​​are obtained at once.

4. The cross-switch chip ACL management method according to claim 1, characterized in that: Methods for inserting ACL configuration into instance blocks include: First, according to the ACL configuration priority, find the location i to be inserted; Secondly, find the first free instance positions b and a before and after the insertion point respectively; Again, compare ib with ai. If ib is smaller, move ib instances before the insertion point i forward in sequence; otherwise, move ai instances after the insertion point i backward in sequence. Finally, the configuration is inserted into the position i to be inserted.

5. The cross-switch chip ACL management method according to claim 4, characterized in that: Methods for reclaiming ACL instance resources include: reclaiming black holes during insertion and reclaiming black holes at a scheduled time. Reclaiming black holes during insertion is a real-time solution, while reclaiming black holes at a scheduled time is an asynchronous solution that ensures black hole convergence. The two solutions are used together to keep black holes within an acceptable range. Reclaiming black holes during insertion is to reduce existing black holes when inserting ACL configurations. Its mechanism adopts the method of inserting ACL configurations into instance blocks. Timed black hole recovery: When the black hole occupancy ratio exceeds the preset threshold, the timer is started. After the timer times out, if the black hole occupancy ratio still exceeds the preset threshold, the recovery process is started. First, the locations of the black holes are obtained in order. Second, the black hole points are filled in order from small to large black hole locations. Finally, in the entire ACL block partition, two processes, recovery of black holes during insertion and recovery of black holes at regular intervals, are performed.

6. A cross-switch chip ACL management system, characterized in that: The cross-switch chip ACL management method as described in any one of claims 1 to 5 is adopted.