Interrupt information processing method, device, system, equipment, medium and product

By adopting the dual-dimensional classification and arbitration strategy of priority and virtual machine identification in multiple virtual airport scenarios, the problem of uneven interrupt processing in the existing technology is solved, and fine control of virtual machines and pressure uniformization is achieved.

CN119025220APending Publication Date: 2024-11-26SHANGHAI BIREN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411154978.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In multi-virtual airport scenarios, the prior art is difficult to achieve fine control of virtual machines through prioritization of interrupt information, resulting in uneven interrupt processing and possible backpressure or outbreak situations.

Method used

A two-dimensional classification cache and arbitration strategy based on priority and purpose virtual machine identification is adopted to ensure that interrupt information is accurately stored and allocated in the cache, and fine-grained control is carried out during quorum based on priority and virtual machine status.

Benefits of technology

It realizes fine control and backpressure management of specific virtual machines, evensizes the interrupt processing pressure of each virtual machine, and improves the interrupt control efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119025220A_ABST
    Figure CN119025220A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses an interrupt information processing method, device and system, equipment, a medium and a product. The method comprises the following steps: receiving first interrupt information, wherein the first interrupt information comprises a priority of the first interrupt information and an identifier of a target virtual machine of the first interrupt information; storing the first interrupt information in a cache based on the identifier and the priority; and based on an arbitration strategy associated with at least one of the identifier and the priority, arbitrating an execution state of sending the first interrupt information in the cache to the target virtual machine. The interrupt information is classified and cached based on the two dimensions of the priority parameter and the identifier of the target virtual machine, and the two dimensions are independently adopted or combined in the arbitration stage to realize arbitration, so that the specific virtual machine can be finely controlled; and the back pressure control of the specific virtual machine can be realized, and the interrupt processing pressure of each virtual machine is more uniform.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of interrupt control, and more specifically, to a method, device, system, equipment, medium and product for processing interrupt information. Background Art

[0002] Interrupt is a mechanism in which when an abnormality or business event occurs inside a device, the device notifies the host of the interruption information so that the host can handle the event in a timely manner. The interrupt mechanism ensures that the device can be handled promptly and effectively when encountering problems, avoiding possible business losses. In modern computer systems and network devices, the interrupt handling mechanism is widely used to ensure the stable operation of the business.

[0003] Typically, an interrupt controller caches interrupt information by category based on priority parameters. However, when a device (such as a chip) operates in a multi-virtual machine usage scenario, it is difficult to achieve fine control of the virtual machine by only distinguishing interrupt information by priority. Summary of the invention

[0004] The present invention provides a method, device, system, equipment, medium and product for processing interrupt information, which are helpful to improve the fine control of virtual machines in a multi-virtual machine scenario.

[0005] The technical solution of the embodiment of the present invention is as follows:

[0006] A method for processing interrupt information, comprising:

[0007] Receive first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information;

[0008] Based on the identifier and the priority, storing the first interrupt information in a cache;

[0009] Based on an arbitration policy associated with at least one of the identifier and the priority, an execution state of sending the first interrupt information in the cache to the destination virtual machine is arbitrated.

[0010] In one embodiment, the number of priorities is M, where M is a positive integer of at least 2;

[0011] The receiving first interrupt information comprises:

[0012] The first interrupt information is received from an interrupt source of a device, wherein N virtual machines including the destination virtual machine are deployed in the device, and N is a positive integer of at least 2.

[0013] In one implementation, storing the first interrupt information in a cache based on the identifier and the priority includes:

[0014] Determine a cache queue set corresponding to the identifier in the cache, the cache queue set comprising M cache queues corresponding to respective priorities;

[0015] Determine a first cache queue corresponding to the priority from the M cache queues;

[0016] In the first cache queue, the first interrupt information is stored in a first-in-first-out manner.

[0017] In one embodiment, the arbitrating the execution state of sending the first interrupt information in the cache to the destination virtual machine based on the arbitration strategy associated with at least one of the identifier and the priority includes at least one of the following:

[0018] Based on a first arbitration strategy indicating that the interrupt information containing the identifier is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and second interrupt information in a cache queue different from the first cache queue in the cache, containing the identifier and having a priority different from that of the first interrupt information, is preferentially sent to the destination virtual machine of the second interrupt information;

[0019] Based on a second arbitration strategy indicating that the interrupt information containing the priority is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and that the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, is preferentially sent to the destination virtual machine of the third interrupt information;

[0020] Based on a third arbitration strategy indicating that the interrupt information including the identifier and the priority level is sent preferentially, determining that the first interrupt information in the first cache queue is sent preferentially to the destination virtual machine of the first interrupt information;

[0021] Based on a fourth arbitration strategy representing stopping sending the interrupt information containing the identifier, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the second interrupt information in the cache queue different from the first cache queue, containing the identifier and having a priority different from that of the first interrupt information, to the destination virtual machine of the second interrupt information;

[0022] Based on a fifth arbitration strategy representing stopping sending the interrupt information containing the priority, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, to the destination virtual machine of the third interrupt information;

[0023] Based on a sixth arbitration policy indicating stopping sending the interrupt information including the identifier and the priority, it is determined to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information.

[0024] In one embodiment, the method comprises:

[0025] Obtaining a back pressure state of a destination virtual machine of the first interrupt information;

[0026] The arbitrating, based on the arbitration strategy associated with at least one of the identifier and the priority, the execution state of sending the first interrupt information in the cache to the destination virtual machine includes:

[0027] Based on the back pressure state and the arbitration policy, arbitrating an execution state of a destination virtual machine that sends the first interrupt information in the cache to the first interrupt information;

[0028] When the back pressure state represents that the destination virtual machine of the first interrupt information is in a busy state, it is decided to stop sending the cached first interrupt information to the destination virtual machine of the first interrupt information; when the back pressure state represents that the destination virtual machine of the first interrupt information is in an idle state, based on the arbitration strategy, the execution state of sending the cached first interrupt information to the destination virtual machine of the first interrupt information is arbitrated.

[0029] A device for processing interrupt information, comprising:

[0030] A receiving module, configured to receive first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information;

[0031] A cache module, configured to store the first interrupt information in a cache based on the identifier and the priority;

[0032] The arbitration module is used to arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine based on an arbitration strategy associated with at least one of the identifier and the priority.

[0033] In one embodiment, the number of priorities is M, where M is a positive integer of at least 2;

[0034] The receiving module is used to receive the first interrupt information from an interrupt source of a device, wherein N virtual machines including the destination virtual machine are deployed in the device, and N is a positive integer of at least 2.

[0035] In one embodiment, the cache module is used to determine a cache queue set corresponding to the identifier in the cache, the cache queue set including M cache queues corresponding to respective priorities; determine a first cache queue corresponding to the priority from the M cache queues; and store the first interrupt information in the first cache queue in a first-in-first-out manner.

[0036] In one embodiment, the arbitration module is used to perform at least one of the following:

[0037] Based on a first arbitration strategy indicating that the interrupt information containing the identifier is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and second interrupt information in a cache queue different from the first cache queue in the cache, containing the identifier and having a priority different from that of the first interrupt information, is preferentially sent to the destination virtual machine of the second interrupt information;

[0038] Based on a second arbitration strategy indicating that the interrupt information containing the priority is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and that the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, is preferentially sent to the destination virtual machine of the third interrupt information;

[0039] Based on a third arbitration strategy indicating that the interrupt information including the identifier and the priority level is sent preferentially, determining that the first interrupt information in the first cache queue is sent preferentially to the destination virtual machine of the first interrupt information;

[0040] Based on a fourth arbitration strategy representing stopping sending the interrupt information containing the identifier, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the second interrupt information in the cache queue different from the first cache queue, containing the identifier and having a priority different from that of the first interrupt information, to the destination virtual machine of the second interrupt information;

[0041] Based on a fifth arbitration strategy representing stopping sending the interrupt information containing the priority, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, to the destination virtual machine of the third interrupt information;

[0042] Based on a sixth arbitration policy indicating stopping sending the interrupt information including the identifier and the priority, it is determined to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information.

[0043] In one embodiment, the arbitration module is used to obtain the back pressure status of the destination virtual machine of the first interrupt information; based on the back pressure status and the arbitration strategy, arbitrate the execution status of sending the first interrupt information in the cache to the destination virtual machine of the first interrupt information; when the back pressure status indicates that the destination virtual machine of the first interrupt information is in a busy state, it is decided to stop sending the cached first interrupt information to the destination virtual machine of the first interrupt information; when the back pressure status indicates that the destination virtual machine of the first interrupt information is in an idle state, based on the arbitration strategy, arbitrate the execution status of sending the cached first interrupt information to the destination virtual machine of the first interrupt information.

[0044] A system for processing interrupt information includes an interrupt source, N virtual machines and an interrupt controller, where N is a positive integer of at least 2;

[0045] The interrupt source is used to generate first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information among the N virtual machines;

[0046] The interrupt controller is used to store the first interrupt information in a cache based on the identifier and the priority; and arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine based on an arbitration strategy associated with at least one of the identifier and the priority.

[0047] In one embodiment, the interrupt source includes at least one of the following:

[0048] Direct memory access controller; video encoder; video decoder; audio encoder; audio decoder; image encoder; image decoder.

[0049] An electronic device, comprising:

[0050] Memory;

[0051] processor;

[0052] The memory stores an application program executable by the processor, which is used to enable the processor to execute any of the above-mentioned methods for processing interrupt information.

[0053] A computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the processor executes any of the above-mentioned methods for processing interrupt information.

[0054] A program product includes a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for processing interrupt information.

[0055] It can be seen from the above technical solution that in an embodiment of the present invention, a first interrupt message is received, the first interrupt message includes the priority of the first interrupt message and the identifier of the destination virtual machine of the first interrupt message; based on the identifier and the priority, the first interrupt message is stored in a cache; based on an arbitration strategy associated with at least one of the identifier of the destination virtual machine and the priority of the first interrupt message, the execution state of sending the first interrupt message in the cache to the destination virtual machine is arbitrated. It can be seen that the embodiment of the present invention classifies and caches the interrupt message based on the two dimensions of the priority parameter and the identifier of the destination virtual machine of the interrupt message, and adopts these two dimensions alone or in combination to achieve arbitration in the arbitration stage, which can finely control a specific virtual machine, and can also achieve back pressure control of a specific virtual machine and make the interrupt processing pressure of each virtual machine more uniform. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of caching interrupt information based on priority in the related art.

[0057] Figure 2 is an exemplary flow chart of a method for processing interrupt information according to an embodiment of the present invention.

[0058] Figure 3 It is a schematic diagram of caching interruption information based on priority and destination virtual machine identifier according to an embodiment of the present invention.

[0059] Figure 4 is an exemplary structural diagram of an interrupt processing system according to an embodiment of the present invention.

[0060] Figure 5 is an exemplary structural diagram of an interrupt controller according to an embodiment of the present invention.

[0061] Figure 6 FIG. 4 is an exemplary schematic diagram of an interrupt selection module according to an embodiment of the present invention.

[0062] Figure 7 is an exemplary schematic diagram of an interrupt arbitration module according to an embodiment of the present invention.

[0063] Figure 8 FIG. 4 is an exemplary structural diagram of an interrupt information processing device according to an embodiment of the present invention.

[0064] Fig. 9 is an exemplary structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings.

[0066] For the sake of brevity and intuitiveness in description, the scheme of the present invention is explained below by describing several representative implementations. A large number of details in the implementations are only used to help understand the scheme of the present invention. However, it is obvious that the technical scheme of the present invention may not be limited to these details when implemented. In order to avoid unnecessarily obscuring the scheme of the present invention, some implementations are not described in detail, but only a framework is given. Hereinafter, "including" means "including but not limited to", and "according to..." means "at least according to..., but not limited to only according to...". Due to the language habits of Chinese, when the number of a component is not specifically specified below, it means that the component can be one or more, or can be understood as at least one.

[0067] In the related art, the interrupt controller classifies and caches the interrupt information based on a single dimension, which is a priority parameter.

[0068] Figure 1 The figure is a schematic diagram of caching interrupt information based on priority in the related art. In the interrupt selection stage, the priority contained in the interrupt information sent by the interrupt source is identified, and the interrupt information is placed in the corresponding queue in the cache based on the identified priority. For example, when the priority identified from interrupt information 1 is 1, interrupt information 1 is placed in the cache queue corresponding to priority 1 in the cache; when the priority identified from interrupt information 2 is 3, interrupt information 2 is placed in the cache queue corresponding to priority 3 in the cache.

[0069] In the interrupt arbitration phase, based on the arbitration policy associated with the priority, the cache queues of all priorities in the cache are arbitrated to be sent to the virtual machine. For example, assuming that the arbitration policy is to give priority to sending interrupt information with a priority of 4, the interrupt information with a priority of 4 is sent in a first-in-first-out order from the cache queue corresponding to the priority of 4 in the cache.

[0070] However, when a device (such as a chip) works in a scenario with multiple virtual machines, this method of caching and arbitrating interrupt information through a single dimension of priority makes it difficult to achieve fine control for each virtual machine. Moreover, in a scenario with multiple virtual machines, if multiple interrupt information with destination addresses of different virtual machines are mixed in the same cache queue in the cache as in the related art, when any virtual machine is back-pressured due to a busy state or other reasons, the interrupt information of other virtual machines cannot be decoupled in the cache, which will cause back-pressure on the interrupt information of all virtual machines, thereby affecting the efficiency of interrupt control. In addition, the related art method of not considering the virtual machine in the interrupt arbitration stage may lead to uneven interrupt arbitration between virtual machines, and interrupt backlogs or bursts within the virtual machine.

[0071] In an embodiment of the present invention, taking into account the special requirements in the usage scenario of multiple virtual machines, the interrupt information is not classified and cached based solely on the single dimension of the priority parameter, but is classified and cached based on the two dimensions of the priority parameter and the identifier of the destination virtual machine of the interrupt information, and these two dimensions are used separately or in combination to achieve arbitration in the arbitration stage, which can finely control a specific virtual machine, realize back pressure control of a specific virtual machine, and make the interrupt processing pressure of each virtual machine more uniform.

[0072] The above disclosure details the technical defects existing in the related art, the causes of the technical defects, and the thinking and analysis process of overcoming the technical defects. In fact, the cognition of the above technical defects is not common knowledge in the field, but a novel discovery made by the applicant in research. In addition, the cause tracing of the technical defects and the thinking and analysis process of overcoming the technical defects are also the gradual analysis results of the applicant in the actual research process, and are not common knowledge in the field.

[0073] Figure 2 is an exemplary flow chart of a method for processing interrupt information according to an embodiment of the present invention. Figure 2 The process shown may be executed by a device such as an interrupt controller. Figure 2 As shown, the method includes:

[0074] Step 101: Receive first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information.

[0075] Here, the first interrupt information not only includes the priority of the first interrupt information, but also includes the identifier of the destination virtual machine of the first interrupt information. Among them, the identifier of the destination virtual machine is preferably unique. The destination virtual machine of the first interrupt information is the virtual machine to which the first interrupt information is sent. After receiving the first interrupt information, the destination virtual machine can respond to the first interrupt information to perform one or more reaction actions based on a predetermined logic, or not respond. The number of first interrupt information can be one or more.

[0076] In one embodiment, the number of priorities is M, where M is a positive integer of at least 2; step 101 specifically includes: receiving a first interrupt message from an interrupt source of a device, where N virtual machines including a destination virtual machine are deployed in the device, and N is a positive integer of at least 2. Here, the number of priorities being M means that the total number of priorities contained in all the first interrupt messages is M. For example, assuming that there are 100 first interrupt messages, each of which contains its own priority (different first interrupt messages may have the same priority). Assuming that these 100 interrupt messages contain a total of 10 priorities, namely priority 1, priority 2...priority 10, the number of priorities is 10.

[0077] Typically, the number of priorities contained in a single piece of first interrupt information is 1, and the priority has been specified by the interrupt source.

[0078] An interrupt source is the cause or source of an interruption of a device (e.g., a computer, various types of chips, integrated circuits, or systems on a chip, etc.), that is, an entity that can send an interrupt message (e.g., an interrupt request signal). These entities can be hardware devices, software programs, or other system events that request to suspend the current task and handle these interrupt messages by sending interrupt messages.

[0079] Specifically, the interrupt sources may include: (1) external interrupt sources: for example, triggered by computer external devices, such as input and output devices such as keyboards, printers, mice, and data channels such as networks and disks, etc. (2) internal interrupt sources: for example, CPU internal events, such as operation errors (such as division by 0, overflow, etc.), program debugging requirements (such as single-step execution, breakpoint debugging), and exceptions caused by special operations (such as memory limit violations, page faults, etc.). Preferably, when the device is a chip or an integrated circuit, the interrupt source can be implemented as: direct memory access controller (DMA); video encoder; video decoder; audio encoder; audio decoder; image encoder; image decoder, etc.

[0080] Step 102: Based on the identifier and the priority, store the first interruption information in a cache.

[0081] In one embodiment, based on the identifier and the priority, storing the first interruption information in the cache includes: determining a cache queue set corresponding to the identifier in the cache, the cache queue set including M cache queues corresponding to respective priorities; determining a first cache queue corresponding to the priority from the M cache queues; and storing the first interruption information in the first cache queue in a first-in-first-out manner. It can be seen that first, the M cache queues corresponding to the identifier in the first interruption information are determined from the cache (since the total number of priorities is M, there are a total of M cache queues corresponding to the identifier in the first interruption information), and then the cache queue corresponding to the priority in the first interruption information (referred to as the first cache queue) is determined from the M cache queues, and then the first interruption information is stored in the first cache queue in a first-in-first-out manner.

[0082] Figure 3 Schematic diagram of caching interruption information based on priority and destination virtual machine identifier according to an embodiment of the present invention. Figure 3 In the scenario of N virtual machines and M priorities, the cache contains N*M cache queues. Each cache queue corresponds to a respective virtual machine ID and a respective priority. That is, one virtual machine corresponds to M cache queues in the cache, so N virtual machines correspond to N*M cache queues in the cache.

[0083] In the interrupt selection phase, the priority and the identifier of the destination virtual machine in the interrupt information received from the interrupt source are identified, and the interrupt information is placed in the corresponding queue in the cache based on the identified priority and the identifier of the destination virtual machine.

[0084] Example 1: When the priority identified from the interrupt information 1 is 1 and the identifier of the destination virtual machine is 1, the interrupt information 1 is placed in the cache queue corresponding to the identifier 1 of the destination virtual machine and the priority 1.

[0085] Example 2: When the identifier of the destination virtual machine identified from the interrupt information 2 is 2 and the priority is 3, the interrupt information 2 is placed in the cache queue corresponding to the identifier of the destination virtual machine 2 and the priority 3.

[0086] It can be seen that, considering the special requirements in the use scenario of multiple virtual machines, the implementation method of the present invention no longer classifies and caches interrupt information based on the single dimension of the priority parameter, but classifies and caches interrupt information based on the two dimensions of priority and the identifier of the destination virtual machine of the interrupt information. Therefore, the cache queue in the cache is related to both the priority and the identifier of the destination virtual machine, realizing refined interrupt information caching, laying the foundation for arbitrating the interrupt information that needs to be sent from the cache based on the subsequent multi-arbitration strategy.

[0087] Step 103: Based on an arbitration strategy associated with at least one of the identifier and the priority, arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine.

[0088] Here, based on the predetermined arbitration strategy, the execution state of sending the first interrupt information in the cache to the destination virtual machine can be arbitrated in a variety of ways. Therefore, considering the special requirements in the use scenario of multiple virtual machines, the implementation method of the present invention no longer arbitrates the interrupt information based on the single dimension of priority, but adopts these two dimensions separately or in combination in the arbitration stage to achieve arbitration, which can finely control the specific virtual machine, and can also achieve back pressure control of the specific virtual machine and make the interrupt processing pressure of each virtual machine more uniform.

[0089] Specifically, based on the arbitration strategy associated with at least one of the identifier and the priority, arbitrating the execution state of sending the first interrupt information in the cache to the destination virtual machine includes:

[0090] (1) Based on an arbitration policy associated with an identifier of a destination virtual machine, an execution state of sending the first interrupt information in the cache to the destination virtual machine is arbitrated.

[0091] (2) Based on an arbitration policy associated with the priority in the interrupt information, arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine.

[0092] (3) Arbitrating the execution state of sending the first interrupt information in the cache to the destination virtual machine based on an arbitration policy associated with the identifier of the destination virtual machine and the priority in the interrupt information.

[0093] In one embodiment, step 103 specifically includes the following situations:

[0094] Scenario (1): Based on the first arbitration strategy that characterizes the priority of sending interrupt information containing the identifier of the first interrupt information, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and the second interrupt information in the cache queue different from the first cache queue, which contains the identifier of the destination virtual machine in the first interrupt information and has a priority different from that of the first interrupt information, is preferentially sent to the destination virtual machine of the second interrupt information.

[0095] For example, assume that the identifier of the destination virtual machine of the first interruption information is identifier 1 and the priority is 1. When the first arbitration policy specifies that the interruption information including identifier 1 is sent first, it is determined that:

[0096] (1) The first interrupt information (identified as 1 and having a priority of 1) in the first cache queue is preferentially sent to the destination virtual machine (identified as 1) of the first interrupt information. For example, the first interrupt information is sent from the first cache queue in a first-in-first-out manner.

[0097] (2) The second interrupt information in other cache queues, whose destination virtual machine is identified as ID 1 and whose priority is not 1, is preferentially sent to the destination virtual machine of the second interrupt information (since the destination virtual machine of the second interrupt information is identified as the same as the destination virtual machine of the first interrupt information, the second interrupt information is sent to the destination virtual machine identified as 1). For example, the second interrupt information is sent from other cache queues in a first-in-first-out manner.

[0098] It can be seen that based on the first arbitration strategy, interrupt information with the same identifier can be sent preferentially.

[0099] Scenario (two): Based on the second arbitration strategy that characterizes the priority of sending interrupt information containing the priority of the first interrupt information, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and the third interrupt information in the cache that is different from the first cache queue, contains the priority of the first interrupt information and the identifier of the destination virtual machine is different from the identifier of the destination virtual machine of the first interrupt information, is preferentially sent to the destination virtual machine of the third interrupt information.

[0100] For example, assume that the identifier of the destination virtual machine of the first interrupt information is identifier 1 and the priority is 1. When the second arbitration policy specifies that the interrupt information containing priority 1 is sent first, it is determined that:

[0101] (1) The first interrupt information (identified as 1 and having a priority of 1) in the first cache queue is preferentially sent to the destination virtual machine (identified as 1) of the first interrupt information. For example, the first interrupt information is sent from the first cache queue in a first-in-first-out manner.

[0102] (2) The third interrupt information in other cache queues whose destination virtual machine identifier is not identifier 1 and whose priority is 1 is preferentially sent to the destination virtual machine of the third interrupt information. For example, the third interrupt information is sent from other cache queues in a first-in-first-out manner.

[0103] It can be seen that based on the second arbitration strategy, interrupt information with the same priority level can be sent preferentially.

[0104] Scenario (three): based on the third arbitration policy indicating priority in sending the interrupt information including the identifier of the first interrupt information and the priority of the first interrupt information, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information.

[0105] For example, assume that the identifier of the destination virtual machine of the first interruption information is identifier 1 and the priority is 1. When the third arbitration policy specifies that the interruption information containing identifier 1 and the priority is 1 is sent preferentially, the decision is:

[0106] (1) The first interrupt information (identified as 1 and having a priority of 1) in the first cache queue is preferentially sent to the destination virtual machine (identified as 1) of the first interrupt information. For example, the first interrupt information is sent from the first cache queue in a first-in-first-out manner.

[0107] It can be seen that based on the third arbitration strategy, the interrupt information to be sent can be accurately selected based on the priority and the identifier.

[0108] Scenario (IV): Based on the fourth arbitration strategy that characterizes stopping sending interrupt information containing the identifier of the first interrupt information, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the second interrupt information in the cache queue different from the first cache queue, which contains the identifier of the destination virtual machine in the first interrupt information and has a priority different from the priority of the first interrupt information, to the destination virtual machine of the second interrupt information.

[0109] For example, assume that the identifier of the destination virtual machine of the first interruption information is identifier 1 and the priority is 1. When the fourth arbitration policy specifies to stop sending the interruption information containing identifier 1, it is determined that:

[0110] (1) Stop sending the first interrupt information (identified as 1 and having a priority of 1) in the first cache queue to the destination virtual machine (identified as 1) of the first interrupt information.

[0111] (2) Stop sending the second interrupt information in other cache queues, whose destination virtual machine is identified as 1 and whose priority is not 1, to the destination virtual machine of the second interrupt information (because the identifier of the destination virtual machine of the second interrupt information is equal to the identifier of the destination virtual machine of the first interrupt information, that is, stop sending the second interrupt information to the destination virtual machine identified as 1).

[0112] It can be seen that based on the fourth arbitration strategy, sending interrupt information with the same identifier can be stopped.

[0113] Scenario (V): Based on the fifth arbitration strategy that characterizes stopping sending interrupt information containing the priority of the first interrupt information, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the third interrupt information in the cache queue different from the first cache queue, which contains the priority of the first interrupt information and the identifier of the destination virtual machine is different from the identifier of the destination virtual machine of the first interrupt information, to the destination virtual machine of the third interrupt information.

[0114] For example, assume that the identifier of the destination virtual machine of the first interrupt information is identifier 1 and the priority is 1. When the second arbitration policy specifies to stop sending the interrupt information containing priority 1, it is ruled that:

[0115] (1) Stop sending the first interrupt information (identified as 1 and having a priority of 1) in the first cache queue to the destination virtual machine (identified as 1) of the first interrupt information.

[0116] (2) Stop sending the third interrupt information in other cache queues, whose destination virtual machine identifier is not identifier 1 and whose priority is 1, to the destination virtual machine of the third interrupt information.

[0117] It can be seen that based on the fifth arbitration strategy, the sending of interrupt information with the same priority level can be stopped.

[0118] Scenario (six): based on the sixth arbitration policy indicating stopping sending the interrupt information including the identifier of the destination virtual machine in the first interrupt information and the priority of the first interrupt information, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information.

[0119] For example, assume that the identifier of the destination virtual machine of the first interrupt information is identifier 1 and the priority is 1. When the third arbitration policy specifies to stop sending the interrupt information containing identifier 1 and the priority is 1, it is ruled that:

[0120] (1) Stop sending the first interrupt information (identified as 1 and having a priority of 1) in the first cache queue to the destination virtual machine (identified as 1) of the first interrupt information.

[0121] It can be seen that based on the sixth arbitration strategy, the sending of interrupt information can be accurately stopped based on the priority and the identifier.

[0122] The above exemplary description is of a typical example of implementing arbitration based on an arbitration strategy. Those skilled in the art may appreciate that such description is merely exemplary and is not intended to limit the protection scope of the embodiments of the present invention.

[0123] In one embodiment, the method includes: obtaining the back pressure status of the destination virtual machine of the first interrupt information; arbitrating the execution status of sending the first interrupt information in the cache to the destination virtual machine based on the arbitration strategy associated with the identifier and at least one of the priorities, including: arbitrating the execution status of sending the first interrupt information in the cache to the destination virtual machine of the first interrupt information based on the back pressure status and the arbitration strategy; wherein when the back pressure status indicates that the destination virtual machine of the first interrupt information is in a busy state, deciding to stop sending the cached first interrupt information to the destination virtual machine of the first interrupt information; when the back pressure status indicates that the destination virtual machine of the first interrupt information is in an idle state, arbitrating the execution status of sending the cached first interrupt information to the destination virtual machine of the first interrupt information based on the arbitration strategy.

[0124] For example: Assuming that the destination virtual machine of the first interrupt information receives a large amount of interrupt information and is in a busy state, the destination virtual machine of the first interrupt information provides a back pressure state to the interrupt controller, and the back pressure state indicates that the destination virtual machine is in a busy state. Based on the back pressure state, the interrupt controller decides to stop sending the cached first interrupt information to the destination virtual machine. At this time, arbitration can still be performed normally for other cache queues in the cache. It can be seen that since the interrupt information of each virtual machine has been decoupled based on different cache queues in the cache, the virtual machine in the busy state can be back pressured in a targeted manner (that is, the interrupt information is not sent to the virtual machine in the busy state) without affecting the processing of other normal virtual machines.

[0125] Figure 4 The present invention is an exemplary structural diagram of an interrupt processing system according to an embodiment of the present invention. The interrupt information processing system includes an interrupt source, N virtual machines and an interrupt controller, where N is a positive integer of at least 2; the interrupt source is used to generate first interrupt information, the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information among the N virtual machines; the interrupt controller is used to store the first interrupt information in a cache based on the identifier and the priority; based on an arbitration strategy of at least one of the identifier of the destination virtual machine associated with the first interrupt information and the priority of the first interrupt information, the execution state of sending the first interrupt information in the cache to the destination virtual machine is arbitrated.

[0126] In one embodiment, the interrupt sources include: a direct memory access controller; a video encoder; a video decoder; an audio encoder; an audio decoder; an image encoder; an image decoder, and the like.

[0127] In one embodiment, the interrupt information processing system may be implemented as: a computer, a chip, an integrated circuit or a system on a chip, etc. The chip may include any one of a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a neural network processing unit (NPU), a deep learning processing unit (DPU), an accelerated processing unit (APU), and a general-purpose computing on graphics processing unit (GPGPU).

[0128] In the embodiment of the present invention, the interrupt information is classified in the interrupt controller based on the two dimensions of priority and the identification of the destination virtual machine. The specific circuit structure of the interrupt controller can adopt the traditional three-level sequence of interrupt selection → interrupt cache → interrupt arbitration. By optimizing the internal logic of each module in the interrupt controller, an interrupt controller compatible with multi-virtual machine usage scenarios can be realized.

[0129] Figure 5 FIG. 1 is an exemplary structural diagram of an interrupt controller according to an embodiment of the present invention. Figure 5 In the embodiment, the interrupt controller includes: a bus interface, an interrupt configuration register, an interrupt input selection module, a cache, an interrupt arbitration module and an interrupt output control module. The bus input can be received via the bus interface. By reading the address space accessed by the bus, the bus input is judged to be interrupt information or configuration parameters. The interrupt configuration register stores the configuration parameters from the bus, manages other modules in the interrupt controller, and sends the configuration parameters to the interrupt input selection module. The interrupt input selection module receives the interrupt information and selectively stores the interrupt information in the cache module according to the configuration parameters provided by the interrupt configuration register. The cache module stores the interrupt information and classifies the interrupt information according to the priority and the identifier of the destination virtual machine. The interrupt arbitration module is used to extract the interrupt information from the cache module based on the arbitration strategy associated with at least one of the identifier and the priority of the destination virtual machine when the interrupt output control module is idle. The interrupt output control module is used to connect the output interface and control the sending of the interrupt information.

[0130] In the interrupt selection stage, the interrupt input selection module identifies the content of the input interrupt information (for example, the priority of the interrupt information and the identifier of the destination virtual machine of the interrupt information) according to the configuration parameters, and places the interrupt information into the cache queue corresponding to the content. The priority distinction can meet the latency requirements of different tasks. For more urgent interrupt tasks, the priority can be used to achieve the purpose of sending the last one generated first. In the cache, multiple independent cache queues are distinguished according to the priority and the identifier of the destination virtual machine. Each cache queue can be implemented by a first-in-first-out (First Input First Output, FIFO) structure. In the interrupt arbitration stage, the interrupt arbitration module reads the status of each cache queue and the content of the interrupt information, and implements arbitration based on a predetermined arbitration strategy (such as high priority output first).

[0131] The following takes the example of supporting a maximum of two virtual machines to run simultaneously and dividing the interrupt priority into four levels as an example to illustrate the interrupt selection module and the interrupt arbitration module.

[0132] Figure 6 It is an exemplary schematic diagram of an interrupt selection module according to an embodiment of the present invention. When the interrupt information is transmitted to the interrupt selection module, the interrupt selection module matches the interrupt information with the queue enable signal (queue_enable) provided by the interrupt configuration register to determine the validity of the interrupt information. After confirming the validity of the interrupt information, an interrupt sampling process is performed to parse the priority and the identifier of the virtual machine from the interrupt information, and the specific queue for performing write control in the cache is determined based on the priority and the identifier of the virtual machine, so that the interrupt information is written to the queue in the cache. In the process of transmitting the interrupt information from the interrupt selection module to the cache, an overflow check mechanism is supported to avoid overflow of the interrupt information.

[0133] In the cache, each cache queue can adopt a FIFO structure. Multiple interrupt information in the same cache queue adopts the strategy of first input first consumed by the interrupt arbitration module. Multiple cache queues are executed in parallel, and the control bits and status information of each cache queue are presented to the interrupt selection module and the interrupt arbitration module to achieve independent control in the two dimensions of priority and virtual machine identification. In the use scenario of multiple virtual machines, the waiting status of the cache queue usually includes: (1) the valid indication signal of the interrupt information generated by the interrupt cache module and divided by queue granularity; (2) the back pressure of each virtual machine on the interrupt controller from the downstream virtual machine; (3) the stop sending (VF_disable) signal from the interrupt control register, which controls the interrupt controller to stop arbitrating the interrupt information of the corresponding virtual machine queue, thereby realizing the interrupt stop sending control at the virtual machine granularity.

[0134] Figure 7It is an exemplary schematic diagram of an interrupt arbitration module according to an embodiment of the present invention. The interrupt arbitration module can adopt an arbitration strategy of first priority and then polling based on identification. When 4 priority divisions are supported, the interrupt information with the highest priority can be arbitrated. The interrupt information of each virtual machine within the same priority can be evenly arbitrated in a polling manner. The interrupt arbitration module also supports a priority sending function based on virtual machines. By controlling the virtual machine control signal (VF_mask_control) in the interrupt configuration register to distinguish whether there is a virtual machine that needs to send an interrupt first. In the priority scheduling stage, the interrupt arbitration module gives priority to collecting the validity of the interrupt information of the virtual machine controlled by VF_mask_control, and then performs polling scheduling to meet the flexible interrupt control requirements of the system.

[0135] For example: In an environment with 4 priorities and 2 virtual machines, it is assumed that interrupt information is to be sent in cache queue 1 of virtual machine 0 and priority 0 and cache queue 2 of virtual machine 0 and priority 1, and interrupt information is to be sent in cache queue 3 of virtual machine 1 and priority 2. The host configures the VF_mask_control signal to control the interrupt information of virtual machine 1 to be sent first. The interrupt arbitration module collects the status of each cache queue and the VF_mask_control signal. The queue status is stored in two branches: the normal sending branch of the virtual machine and the priority sending branch of the virtual machine. In the normal sending branch of the virtual machine: the interrupt arbitration module collects all interrupt status of virtual machine 0 and virtual machine 1, and priority 0, priority 1 and priority 2 are all judged as active. In the priority sending branch of the virtual machine, since VF_mask_control indicates that the interrupt information of virtual machine 1 is sent first, the interrupt arbitration module only determines mask_priority_2 as active. There is an active queue in the priority sending branch of the virtual machine, and the interrupt arbitration module arbitrates this branch first. After the selection of polling scheduling, it is finally determined to send the interrupt information of virtual machine 1 and priority 2 among the three types of interrupt information to be sent.

[0136] In summary, in an embodiment of the present invention, first interrupt information is received, the first interrupt information includes the priority of the first interrupt information and the identifier of the destination virtual machine of the first interrupt information; based on the identifier and the priority, the first interrupt information is stored in a cache; based on an arbitration strategy associated with at least one of the identifier and the priority, the execution state of sending the first interrupt information in the cache to the destination virtual machine is arbitrated. It can be seen that the embodiment of the present invention classifies and caches interrupt information based on the two dimensions of priority parameter and the identifier of the destination virtual machine of the interrupt information, and adopts these two dimensions alone or in combination to achieve arbitration in the arbitration stage, which can finely control a specific virtual machine, and can also achieve back pressure control of a specific virtual machine and make the interrupt processing pressure of each virtual machine more uniform.

[0137] Figure 8 FIG. 1 is an exemplary structural diagram of an interrupt information processing device according to an embodiment of the present invention. Figure 8 As shown, the interrupt information processing device 300 includes: a receiving module 301, used to receive the first interrupt information, the first interrupt information includes the priority of the first interrupt information and the identifier of the destination virtual machine of the first interrupt information; a cache module 302, used to store the first interrupt information in the cache based on the identifier and the priority; an arbitration module 303, used to arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine based on an arbitration strategy associated with at least one of the identifier and the priority.

[0138] In one embodiment, the number of priorities is M, where M is a positive integer of at least 2; the receiving module 301 is used to receive first interrupt information from an interrupt source of the device, where N virtual machines including the destination virtual machine are deployed in the device, and N is a positive integer of at least 2.

[0139] In one embodiment, the cache module 302 is used to determine a cache queue set corresponding to the identifier in the cache, the cache queue set including M cache queues corresponding to respective priorities; determine a first cache queue corresponding to the priority from the M cache queues; and store the first interrupt information in the first cache queue in a first-in-first-out manner.

[0140] In one embodiment, the arbitration module 303 is used to perform at least one of the following: based on a first arbitration strategy characterized by giving priority to sending interrupt information containing an identifier, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and the second interrupt information in the cache queue different from the first cache queue, which contains an identifier and has a priority different from the priority of the first interrupt information, is preferentially sent to the destination virtual machine of the second interrupt information; based on a second arbitration strategy characterized by giving priority to sending interrupt information containing a priority, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and the third interrupt information in the cache queue different from the first cache queue, which contains a priority and the identifier of the destination virtual machine is different from the identifier of the first interrupt information, is preferentially sent to the destination virtual machine of the third interrupt information; based on a third arbitration strategy characterized by giving priority to sending interrupt information containing an identifier and a priority, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information. The method comprises the following steps: sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information; based on the fourth arbitration strategy that characterizes stopping sending the interrupt information containing the identifier, deciding to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and stopping sending the second interrupt information in the cache queue different from the first cache queue, containing the identifier and having a priority different from the priority of the first interrupt information, to the destination virtual machine of the second interrupt information; based on the fifth arbitration strategy that characterizes stopping sending the interrupt information containing the priority, deciding to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and stopping sending the third interrupt information in the cache queue different from the first cache queue, containing the priority and the identifier of the destination virtual machine different from the identifier of the first interrupt information, to the destination virtual machine of the third interrupt information; based on the sixth arbitration strategy that characterizes stopping sending the interrupt information containing the identifier and the priority, deciding to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information.

[0141] In one embodiment, the arbitration module 303 is used to obtain the back pressure status of the destination virtual machine of the first interrupt information; based on the back pressure status and the arbitration strategy, arbitrate the execution status of sending the first interrupt information in the cache to the destination virtual machine of the first interrupt information; when the back pressure status indicates that the destination virtual machine of the first interrupt information is in a busy state, it is decided to stop sending the cached first interrupt information to the destination virtual machine of the first interrupt information; when the back pressure status indicates that the destination virtual machine of the first interrupt information is in an idle state, based on the arbitration strategy, arbitrate the execution status of sending the cached first interrupt information to the destination virtual machine of the first interrupt information.

[0142] The embodiment of the present invention further provides an electronic device having a processor-memory architecture. Fig. 9 is a structural diagram of an electronic device according to an embodiment of the present invention. Fig. 9 As shown, the electronic device includes a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401. When the computer program is executed by the processor 401, any of the above interrupt information processing methods is implemented. Among them, the memory 402 can be specifically implemented as a variety of storage media such as an electrically erasable programmable read-only memory (EEPROM), a flash memory (Flash memory), and a programmable program read-only memory (PROM). The processor 401 can be implemented as including one or more central processing units or one or more field programmable gate arrays, wherein the field programmable gate array integrates one or more central processing unit cores. Specifically, the central processing unit or the central processing unit core can be implemented as a CPU, a GPU, a GPGPU, an MCU or a DSP, and the like.

[0143] It should be noted that not all steps and modules in the above processes and structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The division of each module is only for the convenience of describing the functional division adopted. In actual implementation, a module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located in the same device or in different devices.

[0144] The hardware modules in each embodiment can be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuits or logic devices (such as dedicated processors, such as FPGA or ASIC) to perform specific operations. For example, specific operations can be performed in various types of chips (for example, artificial intelligence chips). The hardware module may also include programmable logic devices or circuits (such as general-purpose processors or other programmable processors) temporarily configured by software to perform specific operations. As for whether to implement the hardware module by mechanical means, or by using a dedicated permanent circuit, or by using a temporarily configured circuit (such as configured by software), it can be decided based on cost and time considerations.

[0145] The present invention also provides a machine-readable storage medium, storing instructions for making a machine perform a method as described in the present application. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code for realizing the function of any of the embodiments in the above-mentioned embodiments is stored, and the computer (or CPU or MPU) of the system or device reads out and executes the program code stored in the storage medium. In addition, the operating system etc. operated on the computer can also be completed part or all of the actual operation by instructions based on the program code. The program code read out from the storage medium can also be written to the memory provided in the expansion board inserted into the computer or to the memory provided in the expansion unit connected to the computer, and then the CPU etc. installed on the expansion board or the expansion unit are made to perform part and all of the actual operation based on the instructions of the program code, so as to realize the function of any of the embodiments in the above-mentioned embodiments. The storage medium implementation for providing the program code includes a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card and a ROM. Alternatively, the program code can be downloaded from a server computer or a cloud by a communication network.

[0146] In this article, "schematic" means "serving as an example, instance or explanation", and any diagram or implementation method described as "schematic" in this article should not be interpreted as a more preferred or more advantageous technical solution. In order to make the drawings concise, only the parts related to the present invention are schematically shown in each figure, and do not represent the actual structure of the product. In addition, in order to make the drawings concise and easy to understand, in some figures, only one of the parts with the same structure or function is schematically drawn, or only one of them is marked. In this article, "one" does not mean that the number of the relevant parts of the present invention is limited to "only one", and "one" does not mean that the number of the relevant parts of the present invention is "more than one". In this article, "upper", "lower", "front", "back", "left", "right", "inside", "outside", etc. are only used to indicate the relative position relationship between the relevant parts, rather than to limit the absolute position of these relevant parts.

[0147] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing interrupt information, characterized in that: include: Receive first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information; Based on the identifier and the priority, storing the first interrupt information in a cache; Based on an arbitration policy associated with at least one of the identifier and the priority, an execution state of sending the first interrupt information in the cache to the destination virtual machine is arbitrated.

2. The method according to claim 1, characterized in that The number of priorities is M, where M is a positive integer of at least 2; The receiving first interrupt information comprises: The first interrupt information is received from an interrupt source of a device, wherein N virtual machines including the destination virtual machine are deployed in the device, and N is a positive integer of at least 2.

3. The method according to claim 2, characterized in that The storing the first interrupt information in a cache based on the identifier and the priority includes: Determine a cache queue set corresponding to the identifier in the cache, the cache queue set comprising M cache queues corresponding to respective priorities; Determine a first cache queue corresponding to the priority from the M cache queues; In the first cache queue, the first interrupt information is stored in a first-in-first-out manner.

4. The method according to claim 3, characterized in that The arbitrating, based on the arbitration strategy associated with at least one of the identifier and the priority, the execution state of sending the first interrupt information in the cache to the destination virtual machine includes at least one of the following: Based on a first arbitration strategy indicating that the interrupt information containing the identifier is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and second interrupt information in a cache queue different from the first cache queue in the cache, containing the identifier and having a priority different from that of the first interrupt information, is preferentially sent to the destination virtual machine of the second interrupt information; Based on a second arbitration strategy indicating that the interrupt information containing the priority is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and that the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, is preferentially sent to the destination virtual machine of the third interrupt information; Based on a third arbitration strategy indicating that the interrupt information including the identifier and the priority level is sent preferentially, determining that the first interrupt information in the first cache queue is sent preferentially to the destination virtual machine of the first interrupt information; Based on a fourth arbitration strategy representing stopping sending the interrupt information containing the identifier, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the second interrupt information in the cache queue different from the first cache queue, containing the identifier and having a priority different from that of the first interrupt information, to the destination virtual machine of the second interrupt information; Based on a fifth arbitration strategy representing stopping sending the interrupt information containing the priority, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, to the destination virtual machine of the third interrupt information; Based on a sixth arbitration policy indicating stopping sending the interrupt information including the identifier and the priority, it is determined to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information.

5. The method according to any one of claims 1 to 4, characterized in that The method comprises: Obtaining a back pressure state of a destination virtual machine of the first interrupt information; The arbitrating, based on the arbitration strategy associated with at least one of the identifier and the priority, the execution state of sending the first interrupt information in the cache to the destination virtual machine includes: Based on the back pressure state and the arbitration policy, arbitrating an execution state of a destination virtual machine that sends the first interrupt information in the cache to the first interrupt information; When the back pressure state represents that the destination virtual machine of the first interrupt information is in a busy state, it is decided to stop sending the cached first interrupt information to the destination virtual machine of the first interrupt information; when the back pressure state represents that the destination virtual machine of the first interrupt information is in an idle state, based on the arbitration strategy, the execution state of sending the cached first interrupt information to the destination virtual machine of the first interrupt information is arbitrated.

6. A device for processing interrupt information, characterized in that: include: A receiving module, configured to receive first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information; A cache module, configured to store the first interrupt information in a cache based on the identifier and the priority; The arbitration module is used to arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine based on an arbitration strategy associated with at least one of the identifier and the priority.

7. The device according to claim 6, characterized in that The number of priorities is M, where M is a positive integer of at least 2; The receiving module is used to receive the first interrupt information from an interrupt source of a device, wherein N virtual machines including the destination virtual machine are deployed in the device, and N is a positive integer of at least 2.

8. The device according to claim 7, characterized in that The cache module is used to determine a cache queue set corresponding to the identifier in the cache, wherein the cache queue set includes M cache queues corresponding to respective priorities; determine a first cache queue corresponding to the priority from the M cache queues; and store the first interrupt information in the first cache queue in a first-in-first-out manner.

9. The device according to claim 8, characterized in that The arbitration module is configured to perform at least one of the following: Based on a first arbitration strategy indicating that the interrupt information containing the identifier is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and second interrupt information in a cache queue different from the first cache queue in the cache, containing the identifier and having a priority different from that of the first interrupt information, is preferentially sent to the destination virtual machine of the second interrupt information; Based on a second arbitration strategy indicating that the interrupt information containing the priority is sent preferentially, it is determined that the first interrupt information in the first cache queue is preferentially sent to the destination virtual machine of the first interrupt information, and that the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, is preferentially sent to the destination virtual machine of the third interrupt information; Based on a third arbitration strategy indicating that the interrupt information including the identifier and the priority level is sent preferentially, determining that the first interrupt information in the first cache queue is sent preferentially to the destination virtual machine of the first interrupt information; Based on a fourth arbitration strategy representing stopping sending the interrupt information containing the identifier, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the second interrupt information in the cache queue different from the first cache queue, containing the identifier and having a priority different from that of the first interrupt information, to the destination virtual machine of the second interrupt information; Based on a fifth arbitration strategy representing stopping sending the interrupt information containing the priority, it is decided to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information and to stop sending the third interrupt information in the cache queue different from the first cache queue, containing the priority and with an identifier of the destination virtual machine different from the identifier of the destination virtual machine of the first interrupt information, to the destination virtual machine of the third interrupt information; Based on a sixth arbitration policy indicating stopping sending the interrupt information including the identifier and the priority, it is determined to stop sending the first interrupt information in the first cache queue to the destination virtual machine of the first interrupt information.

10. The device according to any one of claims 6 to 9, characterized in that: The arbitration module is used to obtain the back pressure state of the destination virtual machine of the first interrupt information; based on the back pressure state and the arbitration strategy, arbitrate the execution state of sending the first interrupt information in the cache to the destination virtual machine of the first interrupt information; wherein when the back pressure state indicates that the destination virtual machine of the first interrupt information is in a busy state, it is decided to stop sending the cached first interrupt information to the destination virtual machine of the first interrupt information; When the back pressure state indicates that the destination virtual machine of the first interrupt information is in an idle state, based on the arbitration strategy, arbitration is performed on the execution state of the cached first interrupt information sent to the destination virtual machine of the first interrupt information.

11. A system for processing interrupt information, characterized in that: including an interrupt source, N virtual machines and an interrupt controller, where N is a positive integer of at least 2; The interrupt source is used to generate first interrupt information, where the first interrupt information includes a priority of the first interrupt information and an identifier of a destination virtual machine of the first interrupt information among the N virtual machines; The interrupt controller is configured to store the first interrupt information in a cache based on the identifier and the priority; Based on an arbitration policy associated with at least one of the identifier and the priority, an execution state of sending the first interrupt information in the cache to the destination virtual machine is arbitrated.

12. The system according to claim 11, characterized in that The interrupt source includes at least one of the following: Direct memory access controller; video encoder; video decoder; audio encoder; audio decoder; image encoder; image decoder.

13. An electronic device, characterized in that: include: Memory; processor; The memory stores an application program executable by the processor, which is used to enable the processor to execute the interrupt information processing method according to any one of claims 1 to 5.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the interrupt information processing method according to any one of claims 1 to 5.

15. A program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for processing interrupt information according to any one of claims 1 to 5 is implemented.