Resource scheduling method and system and electronic equipment

Abstract

The invention discloses a resource scheduling method and system and electronic equipment, and belongs to the technical field of terminals. The resource scheduling method comprises the following steps: determining load information of a working scene according to load feature information and service quality information of the working scene of the electronic equipment; transmitting the load information to a scheduler corresponding to the working scene for processing; and determining processor parameters corresponding to the working scene according to the processed load information.

Description

Technical Field

[0001] This application belongs to the technical field of terminals, and specifically relates to a resource scheduling method, system, and electronic device. Background Art

[0002] Currently, the new series of Dimensity chips are a 1+3+4 architecture of all large-core Central Processing Units (CPUs), that is, 1 super large core, 3 extra-large cores, and 4 large cores. The Snapdragon Extreme Edition chips use the self-developed Oryon CPU architecture of 2+6, that is, 2 super large cores and 6 performance cores. The architecture design of the processor has an evolutionary trend towards heavy-load task applications, but it still cannot distinguish the changing characteristics corresponding to the game scenario and the daily application scenario, and there are still problems of performance overkill or power consumption waste in game scenarios with different loads or application scenarios with different loads, which urgently need to be solved.

[0003] In response to the above problems, existing processing solutions perform scheduling for different platforms and various scenarios to maximize the performance and energy efficiency of the chip. Specifically, one type of processing solution is to allocate multiple cores in the multi-core CPU according to the priorities of different application programs. For example, most cores are preferentially allocated to foreground application programs with higher priorities, and a small number of cores are allocated to background application programs with lower priorities. Another type of processing solution is to perform special processing for specific priority or heavy-load scenarios, such as game scenarios. For the type and scenario characteristics of the game, scheduling processing is carried out one by one. Based on this, a scheduling scheme of configuring and tagging the game scenario for recognition and then binding cores and adjusting frequencies accordingly, and a scheduling scheme of setting specific strategies for fixed scenarios of different games have gradually evolved. At the same time, the method of optimizing the kernel general scheduler, that is, the Completely Fair Scheduler (CFS), according to the application scenario characteristics is adopted to improve the energy efficiency of the performance cores.

[0004] However, for the above-mentioned processor scheduling schemes, most of the scheduling schemes for specific application task scenarios rely on manual resource-specific debugging, which is time-consuming and laborious, not intelligent enough, and only improves a small number of application scenarios; while the scheduling schemes that optimize and improve the general scheduler can never accurately distinguish key task scenarios, and have a high coupling degree in performance release matching. Summary of the Invention

[0005] The purpose of the embodiments of this application is to provide a resource scheduling method, system, and electronic device, which can improve the intelligence, comprehensiveness, and accuracy of processor scheduling.

[0006] In a first aspect, an embodiment of the present application provides a resource scheduling method, which includes: determining the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device; transmitting the load information to the scheduler corresponding to the working scenario for processing; determining the processor parameters corresponding to the working scenario according to the processed load information.

[0007] In a second aspect, an embodiment of the present application provides a resource scheduling system, which includes: a kernel layer, including a load decision module, where the load decision module is used to determine the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device; the load decision module is further used to transmit the load information to the scheduler corresponding to the working scenario for processing; a hardware layer, which is used to determine the processor parameters corresponding to the working scenario according to the processed load information.

[0008] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, and the memory stores a program or instruction that can run on the processor. When the program or instruction is executed by the processor, the steps of the resource scheduling method as in the first aspect are implemented.

[0009] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by the processor, the steps of the resource scheduling method as in the first aspect are implemented.

[0010] In a fifth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface, and the communication interface is coupled to the processor. The processor is used to run a program or instruction to implement the steps of the resource scheduling method as in the first aspect.

[0011] In a sixth aspect, an embodiment of the present application provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the steps of the resource scheduling method as in the first aspect.

[0012] In the resource scheduling method provided by the embodiment of the present application, the load information of the working scenario is determined according to the load characteristic information and service quality information of the working scenario of the electronic device; the load information is transmitted to the scheduler corresponding to the working scenario for processing; the processor parameters corresponding to the working scenario are determined according to the processed load information. Through the above resource scheduling method, based on the load characteristic information and service quality information of the working scenario of the electronic device, the load information of the working scenario is predicted, and the corresponding scheduler is selected to process the load information, and then the processor parameters required for the working scenario are scheduled based on the processed load information. In this way, the load information is predicted and resource scheduling is performed based on the relevant information of the working scenario, without relying on manual debugging, which improves the intelligence, comprehensiveness and accuracy of processor scheduling. Description of the Drawings

[0013] Figure 1 It is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application;

[0014] Figure 2 It is one of the schematic diagrams of the principle of the resource scheduling method provided by the embodiment of the present application;

[0015] Figure 3 It is the second schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0016] Figure 4 It is the third schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0017] Figure 5 It is the fourth schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0018] Figure 6 It is a schematic diagram of the operation interface of the resource scheduling method provided by the embodiment of the present application;

[0019] Figure 7 It is the fifth schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0020] Figure 8 It is the sixth schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0021] Figure 9 It is the seventh schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0022] Figure 10 It is the eighth schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0023] Figure 11 It is the ninth schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0024] Figure 12 It is the tenth schematic diagram of the principle of the resource scheduling method provided by the embodiment of the present application;

[0025] Figure 13 It is a block diagram of the structure of the resource scheduling system provided by the embodiment of the present application;

[0026] Figure 14 It is a block diagram of the structure of the electronic device provided by the embodiment of the present application;

[0027] Figure 15 It is a schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present application. Detailed Embodiments

[0028] Next, the technical solutions in the embodiments of the present application will be clearly described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0029] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same category, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0030] Next, in conjunction with the accompanying drawings, the resource scheduling method, system, and electronic device provided by the embodiments of the present application will be described in detail through specific embodiments and their application scenarios.

[0031] As Figure 1 shown, the embodiments of the present application provide a resource scheduling method, which may include the following S102 to S106:

[0032] S102: Determine the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device.

[0033] The resource scheduling method proposed by the embodiments of the present application is executed by an electronic device, which may specifically be an intelligent electronic device such as a smart phone, a tablet computer, a notebook computer, and a smart watch, and no specific limitation is made here.

[0034] Among them, the above-mentioned working scenario may be a game scenario, a non-game scenario, or a combination of the two, and no specific limitation is made here.

[0035] Further, the service quality (QoS) information is used to indicate the running situation of the electronic device during the processing of the current working scenario, such as the response speed.

[0036] Further, the above-mentioned load information is the predicted actual scenario load of the working scenario.

[0037] Specifically, the system framework of the resource scheduling method proposed by the embodiments of the present application may specifically be as Figure 3As shown, in the resource scheduling method proposed in the embodiment of the present application, the load decision module in the kernel layer can obtain the load characteristic information of the working scenario. On this basis, as Figure 10 shown, the load decision module then predicts the actual scenario load of the working scenario based on the load characteristic information of the working scenario and in combination with the preset quality of service information of the working scenario, and obtains the load information of the working scenario.

[0038] S104: Transmit the load information to the scheduler corresponding to the working scenario for processing.

[0039] Specifically, in the resource scheduling method proposed in the embodiment of the present application, as Figure 3 shown, after the load decision module determines the load information of the working scenario, it will transmit the load information to the scheduler corresponding to the working scenario for processing according to the scenario type of the working scenario.

[0040] In the actual application process, the above scheduler can specifically be a Real Time (RT) scheduler, a Virtual Internet Protocol (VIP) scheduler, a Completely Fair Scheduler, an Energy Aware Scheduling (EAS) scheduler, an extensible scheduler, etc., and no specific limitation is made here.

[0041] S106: Determine the processor parameters corresponding to the working scenario according to the processed load information.

[0042] Specifically, in the resource scheduling method proposed in the embodiment of the present application, as Figure 3 shown, after the scheduler processes the load information of the working scenario, the kernel layer transmits the processed load information to the hardware layer. On this basis, as Figure 8 shown, the hardware layer selects and schedules the processor parameters corresponding to the working scenario, such as the processor core and the processor frequency, based on the processed load information, and updates the priority of the rendering thread of the working scenario.

[0043] The above resource scheduling method provided by the embodiment of the present application determines the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device; transmits the load information to the scheduler corresponding to the working scenario for processing; and determines the processor parameters corresponding to the working scenario according to the processed load information. Through the above resource scheduling method, based on the load characteristic information and service quality information of the working scenario of the electronic device, the load information of the working scenario is predicted, and the corresponding scheduler is selected to process the load information, and then the processor parameters required for the working scenario are scheduled based on the processed load information. In this way, the load information is predicted and resource scheduling is performed based on the relevant information of the working scenario, without relying on manual debugging, which improves the intelligence, comprehensiveness and accuracy of processor scheduling.

[0044] In the embodiment of the present application, before the above S102, the above resource scheduling method may specifically further include the following S108 and S110:

[0045] S108: Using the target model, identify the working scenario of the electronic device according to the interface information and running information of the electronic device.

[0046] Among them, as Figure 4 shown, the above target model is obtained by learning and training the task loads and core configuration information of different types of task scenarios for different applications. The target model can record the current foreground tasks and background tasks of the electronic device, identify key task scenarios, and does not need to distinguish fixed task loads through specified processes, and can also distinguish game scenarios and non-game scenarios.

[0047] Specifically, the target model has an Artificial Intelligence Agent (AI Agent). The intelligent agent is a phased key ability of the target model, which has capabilities such as perception, thinking, memory, planning, screen recognition, and execution, and can understand the content on the screen and execute the user's tasks. In the resource scheduling method proposed in the embodiment of the present application, in order to accurately identify the load of key task scenarios, the operating system (OS) side model specifically expands the capabilities of the intelligent agent facing the scenario layer, so that the target model on the intelligent agent side performs local learning. In the actual application process, the training and learning process of the target model for task scenario perception can be specifically as Figure 12 shown to endow the target model with the ability of task scenario perception and recognition.

[0048] Furthermore, the above running information may specifically include the current running load and running core information of the electronic device.

[0049] Specifically, in the resource scheduling method proposed in the embodiment of the present application, asFigure 3 As shown, the scene perception engine module in the application layer uses the target model to identify the working scene of the electronic device according to the interface information and running information of the electronic device. Among them, the scene perception engine module is the key perception module of the intelligent agent and has the ability of the operating system side model.

[0050] S110: Obtain the load characteristic information of the working scene.

[0051] Among them, the load characteristic information is the theoretical scene load of the working scene, and the load characteristic information is determined by the target model based on the task loads of different task scenes in the training and learning process.

[0052] Specifically, in the resource scheduling method proposed in the embodiment of the present application, as Figure 3 shown, after identifying the working scene of the electronic device, the scene perception engine module further obtains the load characteristic information of the working scene based on the identified working scene, and transmits the load characteristic information to the scene perception module in the framework layer. The scene perception module then transmits the received load characteristic information to the load decision module in the kernel layer, so that the load decision module can predict the actual scene load of the working scene and obtain the load information of the working scene.

[0053] In the above embodiments provided by the present application, the target model is used to identify the working scene of the electronic device according to the interface information and running information of the electronic device; obtain the load characteristic information of the working scene. In this way, using the target model to identify the working scene of the electronic device according to the relevant information of the electronic device does not require manual debugging, improves the intelligence of the processor scheduling, reduces the human and material resources consumed by the processor scheduling, and can identify most task scenes and key task scenes based on the target model, improving the comprehensiveness and accuracy of the processor scheduling.

[0054] In the embodiment of the present application, the above S108 may specifically include the following S108a to S108c:

[0055] S108a: Use the target model to capture the display interface of the electronic device and perform interface recognition on the display interface to obtain an interface recognition result.

[0056] Among them, the target model has the ability of screenshot recognition.

[0057] Specifically, in the resource scheduling method proposed in the embodiment of the present application, the scene perception engine module uses the screenshot recognition ability of the target model to capture the display interface of the electronic device and perform interface recognition on the display interface to obtain an interface recognition result.

[0058] S108b: Obtain the running load and running core information of the electronic device.

[0059] Among them, the above-mentioned running load is used to indicate the load of the current running tasks of the electronic device, and the above-mentioned running core information is used to indicate the current running processor core and processor frequency of the electronic device.

[0060] Specifically, in the resource scheduling method proposed in the embodiments of the present application, the scenario awareness engine module will also obtain the current running information of the electronic device, namely the running load and the running core information.

[0061] S108c: Determine the working scenario of the electronic device according to the interface recognition result, the running load and the running core information.

[0062] Specifically, in the resource scheduling method proposed in the embodiments of the present application, after obtaining the interface recognition result, the running load and the running core information, the target model senses and recognizes the current working scenario of the electronic device based on the results of previous learning and training for the task loads and core configuration information of different applications and different types of task scenarios.

[0063] In the above-mentioned embodiments provided by the present application, in the process of using the target model to identify the working scenario of the electronic device according to the interface information and the running information of the electronic device, the target model is used to capture the display interface of the electronic device, and perform interface recognition on the display interface to obtain the interface recognition result; obtain the running load and the running core information of the electronic device; determine the working scenario of the electronic device according to the interface recognition result, the running load and the running core information. In this way, by using the screenshot recognition ability of the target model and combining the running load and the running core information of the electronic device, the working scenario of the electronic device is recognized, without relying on manual debugging, improving the intelligence of scenario recognition, and based on the target model, most task scenarios and key task scenarios can be recognized, improving the comprehensiveness and accuracy of scenario recognition.

[0064] In the embodiments of the present application, the above-mentioned S108a may specifically include the following S112 to S116:

[0065] S112: Perform interface recognition on the display interface through the target model to determine the interface elements in the display interface.

[0066] Among them, the above-mentioned interface elements are user interface (UI) elements in the display interface, such as elements like pictures, buttons, and menus.

[0067] Specifically, in the resource scheduling method proposed in the embodiments of the present application, after capturing the display interface of the electronic device, perform interface recognition on the display interface through the target model to accurately identify the interface elements in the display interface.

[0068] S114: Determine the target area or target element in the display interface according to the position information and semantic information of the interface elements.

[0069] Among them, the above position information is used to indicate the positions of each interface element and their changes, and the position information can be determined by visually positioning the interface elements.

[0070] Furthermore, the above semantic information can be determined by the semantic connection of the interface elements, and the semantic information can specifically include information such as the marking information of the interface elements.

[0071] Specifically, in the resource scheduling method proposed in the embodiments of this application, after identifying the interface elements in the display interface, the target area or target element in the display interface is determined according to the position information and semantic information of the interface elements.

[0072] S116: Determine the interface recognition result according to the target area or target element.

[0073] Specifically, in the resource scheduling method proposed in the embodiments of this application, after identifying the target area or target element in the display interface, the interface recognition result is determined according to the target area or target element to identify the current scene type and window scene of the electronic device, and support understanding and responding to complex user intentions, not limited to simple click operations, but also including operations such as swiping pages.

[0074] It can be understood that in the task scene recognition scheme in the related art, after determining that it is a game application according to the game process or label code control, several fixed scenes are determined for specific games by adapting game interfaces, such as loading scenes, lobby scenes, single-player lane scenes, team battle scenes, and teleportation scenes. The scene recognition is not comprehensive enough and relies on manual debugging.

[0075] In the resource scheduling method proposed in the embodiments of this application, the screenshot recognition ability of the target model is used to sense and recognize the screen interface of the electronic device: through screenshots, UI elements on screens with different resolutions are accurately recognized, and through visual positioning such as the user's operations on pictures, buttons, and menus, and semantic connections such as user markings, key target areas or target elements are accurately recognized to obtain the current scene type and window scene of the electronic device, so as to support understanding and responding to complex user intentions. In this way, it does not rely on manual debugging, improves the intelligence of scene recognition, and can recognize most task scenes and key task scenes, improving the comprehensiveness and accuracy of scene recognition.

[0076] In the above embodiments provided by the present application, in the process of using the target model to capture the display interface of the electronic device and perform interface recognition on the display interface to obtain the interface recognition result, the target model is used to perform interface recognition on the display interface to determine the interface elements in the display interface; according to the position information and semantic information of the interface elements, the target area or target element in the display interface is determined; according to the target area or target element, the interface recognition result is determined. In this way, by using the screenshot recognition ability of the target model, the working scenario of the electronic device is recognized, without relying on manual debugging, the intelligence of scenario recognition is improved, and most task scenarios and key task scenarios can be recognized, improving the comprehensiveness and accuracy of scenario recognition.

[0077] In the embodiment of the present application, the above S104 may specifically include the following S104a and S104b:

[0078] S104a: When the working scenario includes a game scenario, the load information of the game scenario is transmitted to the first scheduler for processing.

[0079] Among them, the first scheduler is the scalable scheduler SCHE_EXT, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario.

[0080] Specifically, in the resource scheduling method proposed in the embodiment of the present application, as Figure 3 shown, after the load decision module determines the load information of the working scenario, for the game scenario, such as Figure 5 the map running scenario, team battle scenario, monster killing scenario, lobby scenario, resurrection scenario, etc. shown, the load decision module transmits the load information of the game scenario to the first scheduler, that is, the scalable scheduler SCHE_EXT, for processing.

[0081] S104b: When the working scenario includes a non-game scenario, the load information of the non-game scenario is transmitted to the second scheduler for processing.

[0082] Among them, the second scheduler may specifically be a VIP scheduler, CFS, RT scheduler, EAS, etc., and no specific limitation is made here.

[0083] Specifically, in the resource scheduling method proposed in the embodiment of the present application, as Figure 3 shown, after the load decision module determines the load information of the working scenario, for the non-game scenario, such as Figure 7 the video scenario, live broadcast scenario, payment scenario, recording scenario, and ticket grabbing scenario, etc. shown, the load decision module transmits the load information of the non-game scenario to the second scheduler for processing.

[0084] It can be understood that in the scheduler solutions in the related art, either the general scheduler such as CFS is locally optimized, or a VIP scheduling solution is added for the load in critical task scenarios, and the coupling degree in performance release matching is relatively high.

[0085] In the resource scheduling method proposed in the embodiments of the present application, for the game scenario, a customized SCHE_EXT scheduler is provided, which can implement a customized kernel driver solution to better allocate the processor performance according to the load of the actual task scenario.

[0086] In the above embodiments provided by the present application, during the process of transmitting the load information to the scheduler corresponding to the working scenario for processing, when the working scenario includes a game scenario, the load information of the game scenario is transmitted to the first scheduler for processing; when the working scenario includes a non-game scenario, the load information of the non-game scenario is transmitted to the second scheduler for processing; wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario. In this way, a customized scheduler for the game scenario can better allocate the processor performance based on the load of the actual task scenario.

[0087] In the embodiments of the present application, the above step of transmitting the load information of the game scenario to the first scheduler for processing may specifically include the following S118 to S122:

[0088] S118: Determine the core configuration information according to the game scenario and the load information.

[0089] Specifically, in the resource scheduling method proposed in the embodiments of the present application, after the load information of the game scenario is transmitted to the first scheduler, the first scheduler determines the corresponding core configuration information according to the specific game scenario and its load information.

[0090] S120: Determine the processor characteristic information according to the core configuration information.

[0091] Specifically, in the resource scheduling method proposed in the embodiments of the present application, after the first scheduler determines the core configuration information corresponding to the game scenario, according to the core configuration information, it determines the processor characteristic information to bind the core configuration information to the relevant processor characteristics.

[0092] S122: Determine the interface driver information of the processor according to the processor characteristic information.

[0093] Specifically, in the resource scheduling method proposed in the embodiments of the present application, after the first scheduler determines the processor characteristic information, it will determine the interface driver information of the processor according to the processor characteristic information, so as to set the processor characteristic information to each processor in the hardware layer through the relevant interfaces of the driver, and achieve better allocation of processor performance according to the load of the actual task scenario.

[0094] Specifically, the first scheduler, namely the SCHE_EXT scheduler, has a scheduling strategy for the load information of the game scenario as Figure 9 shown. By customizing the core in the SCHE_EXT scheduler to implement the Berkeley packet filter scheduler (Bpf scheduler) and the core scheduler Core scheduler, and customizing the corresponding extended scheduling class ext_sched_class and entity in the Core scheduler, a customized scheduling strategy can be implemented according to the load characteristic information of the game scenario.

[0095] Specifically, for the scheduling strategy of the load information of the game scenario, as Figure 9 shown, the game extended scheduling, namely SCHED_EXT_game, realizes the dynamic loading of the scheduling strategy by extending the Berkeley packet filter scheduler program eBPF scheduler program, so as to support writing Berkeley packet filter code, i.e., BPF Code, in the user space. The key logics such as task core selection ops.select_cpu and task distribution scx_bpf_dispatch are registered to the kernel through the interfaces of the struct sched_ext_ops structure. Further, the function pointer of the kernel scheduling class is replaced by the bpf_struct_ops structure mechanism to achieve the switching of the scheduling strategy. Further, the rendering thread of the game scenario is assigned to the global dynamic task queue (Dynamic Sequence Queue, DSQ), and the task is actively pulled by the scene load core configuration corresponding to the perception module. Further, the load scheduling of the task scenario corresponding to the perception module is realized through the priority configuration of the extended scheduling class ext_sched_class, and the existing scheduling strategy is executed for non-game scenarios. Further, in the task core selection ops.select_cpu logic callback, the task is distributed to the load information obtained by the load decision module, and the determination method of the load information is as Figure 10 shown. Further, it supports obtaining the processor frequency that needs to be set through the hook function, i.e., the HOOK function, and updating it to the load information obtained by the load decision module, corresponding to Figure 9Entities in it, such as enqueue_entity which means adding entity, dequeue_entity which means removing entity, and entity_tick which means entity clock cycle, etc.

[0096] In the above embodiments provided by this application, during the process of transmitting the load information of the game scene to the first scheduler for processing, according to the game scene and the load information, the core configuration information is determined; according to the core configuration information, the processor characteristic information is determined; according to the processor characteristic information, the interface driver information of the processor is determined. In this way, by customizing the scheduler for the game scene, the processor performance can be better allocated based on the actual load of the game scene.

[0097] In the embodiments of this application, the processor parameters include the processor core and the processor frequency. The above S106 may specifically include the following S106a:

[0098] S106a: Using the target model, according to the processed load information, processor type, scene type, and the performance information of each cluster of processor cores, determine the target cluster of processor cores and the processor frequency of each processor core in the target cluster of processor cores.

[0099] Among them, the processor type is used to indicate the chip platform corresponding to the processor, such as the Qualcomm Snapdragon processor chip platform Qcom and the MediaTek Dimensity processor chip platform MTK.

[0100] Among them, for the Qualcomm Snapdragon processor core, the latest version of the core is called 2 super cores + 6 performance cores, and the other versions of the core are called 1 large core + 3 medium cores + 4 small cores; for the MediaTek Dimensity processor core, the latest version of the core is called 1 super large core + 3 super large cores + 4 large cores, and the other versions of the core are called 1 large core + 3 medium cores + 4 small cores.

[0101] Specifically, in the resource scheduling method proposed in the embodiments of this application, as Figure 11 shown, the target model is based on the learning and training results of the previous task loads and core configuration information for different applications and different types of task scenarios. According to the processed load information, processor type, scene type, and the performance information of each cluster of processor cores, determine the target cluster of processor cores, and determine the processor frequency of each processor core in the target cluster of processor cores. That is, the target model performs core selection and frequency adjustment processing according to the processed load information, processor type, scene type, and the performance information of each cluster of processor cores.

[0102] For example, for the Qcom high - throughput Snapdragon processor chip platform, in a game scenario, the super - large core is selected and the processor frequency is increased to a high frequency; in a video scenario, the super - large core is selected and the processor frequency is set to a low frequency. For the MTK Dimensity processor chip platform, in a game scenario, the super - large core is selected and the processor frequency is increased to [frequency value not provided in the original], in a video scenario, the super - large core is selected and the processor frequency is set to a medium frequency.

[0103] In the above - mentioned embodiments provided by the present application, the processor parameters include the processor core and the processor frequency. In the process of determining the processor parameters corresponding to the working scenario according to the processed load information, by using the target model, according to the processed load information, the processor type, the scenario type, and the performance information of each cluster of processor cores, the target cluster of processor cores and the processor frequency of each processor core in the target cluster of processor cores are determined. In this way, using the target model for processor scheduling in the task scenario improves the intelligence, comprehensiveness, and accuracy of processor scheduling.

[0104] In the embodiments of the present application, before the above - mentioned S108, the above - mentioned resource scheduling method may specifically further include the following S124 to S132:

[0105] S124: Obtain the first correspondence between the task scenarios of the first type, the task load, and the core configuration information.

[0106] Among them, the first type is the game scenario.

[0107] Specifically, in the resource scheduling method proposed in the embodiments of the present application, for different task scenarios of the first type, obtain the first correspondence between each task scenario, the task load, and the core configuration information, as shown in Table 1 below.

[0108] Table 1: Corresponding table of load and core configuration for game task scenarios

[0109]

[0110]

[0111] S126: Obtain the second correspondence between the task scenarios of the second type, the task load, and the core configuration information.

[0112] Among them, the second type is the non - game scenario.

[0113] Specifically, in the resource scheduling method proposed in the embodiments of the present application, for different task scenarios of the second type, obtain the second correspondence between each task scenario, the task load, and the core configuration information, as shown in Table 2 below.

[0114] Table 2: Corresponding table of load and core configuration for non - game task scenarios

[0115]

[0116]

[0117] S128: Obtain a third correspondence relationship between the task scenarios of the third type, the task load, and the core configuration information.

[0118] Among them, the third type is a multi-task scenario that combines a game scenario and a non-game scenario.

[0119] Specifically, in the resource scheduling method proposed in the embodiments of the present application, for different task scenarios of the third type, obtain a third correspondence relationship between each group of task scenarios, the task load, and the core configuration information, as shown in Table 3 below.

[0120] Table 3: Corresponding Table of Load and Core Configuration for Multi-Task Scenarios

[0121]

[0122]

[0123] For example, as Figure 6 shown, the electronic device is currently in a multi-task scenario of game - voice call, that is, the user plays a game while making a voice call through the electronic device. Among them, for the game scenario, if the processor chip platform is Qcom, the processor core selects the super core, and the processor frequency selects the high frequency; if the processor chip platform is MTK, the processor core selects the super large core, and the processor frequency selects the high frequency. For the voice call scenario, if the processor chip platform is Qcom, the processor core selects the super core, and the processor frequency selects the low frequency; if the processor chip platform is MTK, the processor core selects the super core, and the processor frequency selects the medium frequency.

[0124] S130: Train the target model according to the first correspondence relationship, the second correspondence relationship, and the third correspondence relationship.

[0125] Specifically, in the resource scheduling method proposed in the embodiments of the present application, based on the obtained first correspondence relationship, second correspondence relationship, and third correspondence relationship, learn and train the target model to obtain the trained target model. In this way, as Figure 5 , Figure 7 and Figure 12 shown, for different game scenarios, non-game scenarios, and multi-task scenarios, enable the target model to learn the task load and core configuration information of each task scenario, while endowing the target model with the ability to sense and identify task scenarios, and enabling the target model to perform intelligent scheduling of the processor for each task scenario.

[0126] S132: Update the trained target model according to the feedback information of the electronic device for each task scenario.

[0127] Among them, as Figure 12 shown, the above feedback information can be specifically determined by the carding situation, frame dropping situation, heating situation, and power consumption situation of the electronic device during the operation of the task scenario.

[0128] Specifically, in the resource scheduling method proposed in the embodiments of the present application, during the process of scenario perception and processor scheduling based on the target model, the operation of the electronic device will also be monitored. If there are phenomena such as carding, heating, frame dropping, or abnormal power consumption in the electronic device, based on the operation of the electronic device, obtain the feedback information of the electronic device for the task scenario, and continue to learn and train the target model according to this feedback information. Repeatedly updating the target model in this way improves the processor scheduling performance and resource allocation performance of the target model for different task scenarios.

[0129] Furthermore, during the process of the electronic device running the task scenario, if there are carding and heating phenomena in the electronic device, the target model will also perform corresponding frequency modulation processing and / or core adjustment processing. Specifically, based on the core adjustment and frequency modulation strategy shown in Table 4 below, adjust the current processor core and processor frequency of the task scenario, and adjust the configuration threshold.

[0130] Table 4: Core Selection and Frequency Modulation Strategy Table for Carding and Heating Phenomena

[0131]

[0132]

[0133] For example, during the process of the electronic device running the game-video multi-task scenario, there is a heating phenomenon. If the processor chip platform is MTK, and the current processor core configuration of the electronic device is: for the game scenario, the processor core selects the super large core, and the processor frequency selects the high frequency; for the video scenario, the processor core selects the extra large core, and the processor frequency selects the medium frequency. At this time, the core adjustment and frequency modulation strategy is: for the game scenario, the processor core selects the super large core, and the processor frequency selects the low frequency; for the video scenario, the processor core selects the large core, and the processor frequency selects the low frequency.

[0134] In the above embodiments provided by the present application, before identifying the working scenario of the electronic device based on the interface information and operating information of the electronic device using the target model, the first correspondence relationship between the first type of task scenario and the task load and core configuration information is obtained; the second correspondence relationship between the second type of task scenario and the task load and core configuration information is obtained; the third correspondence relationship between the third type of task scenario and the task load and core configuration information is obtained; the target model is trained according to the first correspondence relationship, the second correspondence relationship, and the third correspondence relationship; and the trained target model is updated according to the feedback information of the electronic device for each task scenario. In this way, the target model can be endowed with scene perception ability and processor intelligent scheduling ability, improving the intelligence, comprehensiveness, and accuracy of processor scheduling.

[0135] In the embodiment of the present application, after the above S106, the above resource scheduling method may specifically further include the following S134:

[0136] S134: Update the load characteristic information of the working scenario according to the feedback information of the electronic device processing the working scenario.

[0137] Specifically, in the resource scheduling method proposed in the embodiment of the present application, during the process of the electronic device processing the working scenario, the running situation of the electronic device is also monitored. On this basis, as Figure 2 shown, if the electronic device has phenomena such as lag and overheating, based on the running situation of the electronic device, the feedback information of the electronic device processing the working scenario is obtained, and according to this feedback information, the load characteristic information of the working scenario is updated to improve the accuracy of subsequent processor scheduling based on the load characteristic information.

[0138] In the above embodiments provided by the present application, after determining the processor parameters corresponding to the working scenario according to the processed load information, the load characteristic information of the working scenario is updated according to the feedback information of the electronic device processing the working scenario. In this way, the accuracy of subsequent processor scheduling based on the load characteristic information is improved.

[0139] In summary, the present application provides a resource scheduling optimization solution that integrates scene sense recognition and a customized scheduler. As Figure 2 shown, the task scenario is directly identified through the OS agent perception module ability, the key task scenario is identified based on the user intention, the load characteristic information of the identified task scenario is obtained, the load prediction is performed based on the load characteristic information, and the frequency modulation and core selection are performed for the game scenario in combination with the customized SCHE_EXT scheduler, and the frequency modulation and core selection are performed for the non-game scenario in combination with the CFS scheduler or the VIP scheduler, providing a high-degree-of-freedom scheduling ability, achieving a balance between performance and energy efficiency, enabling both the game scenario and the non-game scenario to obtain better processor performance and energy efficiency release, and at the same time decoupling the scheduling policy from the traditional kernel scheduling policy.

[0140] That is, the resource scheduling optimization solution provided by this application integrates an agent-aware regularized scenario load recognition mechanism and has the ability of a scheduler for customized game scenarios. Based on this resource scheduling optimization solution, the recognition of task scenarios is more intelligent, flexible, efficient, and extensive: the recognition of task scenarios is more intelligent, no longer relying on fixed and specific task scenarios for code processing, using the current edge-side model capabilities to achieve the perception and recognition of task scenarios, which is more efficient and practical, and the application of capabilities is more flexible and extensive, no longer limited to recognizing a small number of task scenarios, and no longer relying on time-consuming and laborious manual specific processing.

[0141] Furthermore, compared with the CFS scheduling policy, due to the defects of being unable to distinguish game scenarios from non-game scenarios and unable to recognize user intentions, resulting in problems such as low chip performance and energy efficiency and high coupling degree. The scheduling policy of the customized scheduler based on game scenarios in this application decouples the problems of the CFS scheduling policy, can better match the load of task scenarios, sinks the driving customization to the capabilities of the chip platform, and better solves problems such as performance overkill and high energy consumption of the chip platform.

[0142] In the resource scheduling method provided by the embodiments of this application, the execution subject can be a resource scheduling system. In the embodiments of this application, taking the resource scheduling system executing the above resource scheduling method as an example, the resource scheduling system provided by the embodiments of this application is described.

[0143] As Figure 13 shown, the embodiments of this application provide a resource scheduling system 500, which can include the following kernel layer 506 and hardware layer 510. Among them, the kernel layer 506 includes a load decision module 508.

[0144] The kernel layer 506 includes a load decision module 508. The load decision module 508 is used to determine the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device;

[0145] The load decision module 508 is also used to transfer the load information to the scheduler corresponding to the working scenario for processing;

[0146] The hardware layer 510 is used to determine the processor parameters corresponding to the working scenario according to the processed load information.

[0147] The resource scheduling system 500 provided by the embodiments of the present application determines the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device; transmits the load information to the scheduler corresponding to the working scenario for processing; and determines the processor parameters corresponding to the working scenario according to the processed load information. Through the above resource scheduling system 500, based on the load characteristic information and service quality information of the working scenario of the electronic device, the load information of the working scenario is predicted, and the corresponding scheduler is selected to process the load information, and then the processor parameters required for the working scenario are scheduled based on the processed load information. In this way, the load information is predicted and resource scheduling is performed based on the relevant information of the working scenario, without relying on manual debugging, improving the intelligence, comprehensiveness and accuracy of processor scheduling.

[0148] In the embodiments of the present application, as Figure 13 shown, the resource scheduling system 500 further includes: an application layer 502, including a scenario awareness engine module 504, where the scenario awareness engine module 504 is used to use a target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device; the scenario awareness engine module 504 is further used to obtain the load characteristic information of the working scenario.

[0149] The above embodiments provided by the present application use a target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device; obtain the load characteristic information of the working scenario. In this way, using the target model to identify the working scenario of the electronic device according to the relevant information of the electronic device, without relying on manual debugging, improves the intelligence of processor scheduling, reduces the human and material resources consumed by processor scheduling, and can identify most task scenarios and key task scenarios based on the target model, improving the comprehensiveness and accuracy of processor scheduling.

[0150] In the embodiments of the present application, the scenario awareness engine module 504 is specifically used to: use a target model to intercept the display interface of the electronic device, perform interface recognition on the display interface to obtain an interface recognition result; obtain the running load and running core information of the electronic device; and determine the working scenario of the electronic device according to the interface recognition result, running load and running core information.

[0151] In the above embodiments provided by the present application, in the process of using the target model to identify the working scenario of the electronic device according to the interface information and operating information of the electronic device, the target model is used to capture the display interface of the electronic device, and interface recognition is performed on the display interface to obtain an interface recognition result; the operating load and operating core information of the electronic device are obtained; according to the interface recognition result, the operating load and the operating core information, the working scenario of the electronic device is determined. In this way, based on the screenshot recognition ability of the target model, combined with the operating load and operating core information of the electronic device, the working scenario of the electronic device is recognized, without relying on manual debugging, improving the intelligence of scenario recognition, and based on the target model, most task scenarios and key task scenarios can be recognized, improving the comprehensiveness and accuracy of scenario recognition.

[0152] In the embodiment of the present application, the scenario perception engine module 504 is specifically configured to: perform interface recognition on the display interface by intercepting through the target model to determine the interface elements in the display interface; determine the target area or target element in the display interface according to the position information and semantic information of the interface elements; determine the interface recognition result according to the target area or target element.

[0153] In the above embodiments provided by the present application, in the process of using the target model to capture the display interface of the electronic device and perform interface recognition on the display interface to obtain an interface recognition result, interface recognition is performed on the display interface through the target model to determine the interface elements in the display interface; determine the target area or target element in the display interface according to the position information and semantic information of the interface elements; determine the interface recognition result according to the target area or target element. In this way, using the screenshot recognition ability of the target model to recognize the working scenario of the electronic device, without relying on manual debugging, improves the intelligence of scenario recognition, and can recognize most task scenarios and key task scenarios, improving the comprehensiveness and accuracy of scenario recognition.

[0154] In the embodiment of the present application, the load decision module 508 is specifically configured to: in the case that the working scenario includes a game scenario, transmit the load information of the game scenario to the first scheduler for processing; in the case that the working scenario includes a non-game scenario, transmit the load information of the non-game scenario to the second scheduler for processing; wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario.

[0155] In the above embodiments provided by the present application, during the process of transmitting the load information to the scheduler corresponding to the working scenario for processing, when the working scenario includes a game scenario, the load information of the game scenario is transmitted to the first scheduler for processing; when the working scenario includes a non-game scenario, the load information of the non-game scenario is transmitted to the second scheduler for processing; wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario. In this way, customizing the scheduler for the game scenario can better allocate the processor performance based on the load of the actual task scenario.

[0156] In the embodiment of the present application, the load decision module 508 is specifically configured to: determine the core configuration information according to the game scenario and the load information; determine the processor characteristic information according to the core configuration information; and determine the interface driver information of the processor according to the processor characteristic information.

[0157] In the above embodiments provided by the present application, during the process of transmitting the load information of the game scenario to the first scheduler for processing, the core configuration information is determined according to the game scenario and the load information; the processor characteristic information is determined according to the core configuration information; and the interface driver information of the processor is determined according to the processor characteristic information. In this way, customizing the scheduler for the game scenario can better allocate the processor performance based on the load of the actual game scenario.

[0158] In the embodiment of the present application, the processor parameters include the processor core and the processor frequency. The hardware layer 510 is specifically configured to: use the target model to determine the target cluster processor core and the processor frequency of each processor core in the target cluster processor core according to the processed load information, the processor type, the scenario type, and the performance information of each cluster of processor cores.

[0159] In the above embodiments provided by the present application, the processor parameters include the processor core and the processor frequency. During the process of determining the processor parameters corresponding to the working scenario according to the processed load information, use the target model to determine the target cluster processor core and the processor frequency of each processor core in the target cluster processor core according to the processed load information, the processor type, the scenario type, and the performance information of each cluster of processor cores. In this way, based on the target model for processor scheduling of the task scenario, the intelligence, comprehensiveness, and accuracy of processor scheduling are improved.

[0160] In the embodiments of the present application, before using the target model to identify the working scenario of the electronic device based on the interface information and operating information of the electronic device, the scenario perception engine module 504 is further configured to: obtain a first correspondence between the first type of task scenario and the task load and core configuration information; obtain a second correspondence between the second type of task scenario and the task load and core configuration information; obtain a third correspondence between the third type of task scenario and the task load and core configuration information; train the target model according to the first correspondence, the second correspondence, and the third correspondence; update the trained target model according to the feedback information of the electronic device for each task scenario.

[0161] In the above embodiments provided by the present application, before using the target model to identify the working scenario of the electronic device based on the interface information and operating information of the electronic device, obtain a first correspondence between the first type of task scenario and the task load and core configuration information; obtain a second correspondence between the second type of task scenario and the task load and core configuration information; obtain a third correspondence between the third type of task scenario and the task load and core configuration information; train the target model according to the first correspondence, the second correspondence, and the third correspondence; update the trained target model according to the feedback information of the electronic device for each task scenario. In this way, the target model can be endowed with scenario perception ability and the processor intelligent scheduling ability, improving the intelligence, comprehensiveness, and accuracy of processor scheduling.

[0162] In the embodiments of the present application, after determining the processor parameters corresponding to the working scenario according to the processed load information, the scenario perception engine module 504 is further configured to: update the load characteristic information of the working scenario according to the feedback information of the electronic device for processing the working scenario.

[0163] In the above embodiments provided by the present application, after determining the processor parameters corresponding to the working scenario according to the processed load information, update the load characteristic information of the working scenario according to the feedback information of the electronic device for processing the working scenario. In this way, the accuracy of subsequent processor scheduling based on the load characteristic information is improved.

[0164] The resource scheduling system 500 in the embodiments of the present application may be an electronic device or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be a terminal or other devices other than terminals. Exemplarily, the electronic device may be a mobile phone, a tablet computer, a laptop computer, a handheld computer, a vehicle-mounted electronic device, a Mobile Internet Device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. It may also be a server, a Network Attached Storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc. The embodiments of the present application do not make specific limitations.

[0165] The resource scheduling system 500 in the embodiments of the present application may be a device with an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems. The embodiments of the present application do not make specific limitations.

[0166] The resource scheduling system 500 provided in the embodiments of the present application can implement Figure 1 each process implemented by the method embodiments. To avoid repetition, it will not be elaborated here.

[0167] Optionally, as Figure 14 shown, the embodiments of the present application further provide an electronic device 600, including a processor 602 and a memory 604. A program or instruction that can run on the processor 602 is stored on the memory 604. When the program or instruction is executed by the processor 602, it implements each step of the above-mentioned resource scheduling method embodiments and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0168] It should be noted that the electronic devices in the embodiments of the present application include the above-mentioned mobile electronic devices and non-mobile electronic devices.

[0169] Figure 15 A schematic diagram of the hardware structure of an electronic device for implementing the embodiments of the present application.

[0170] The electronic device 700 includes, but is not limited to, components such as a radio frequency unit 701, a network module 702, an audio output unit 703, an input unit 704, a sensor 705, a display unit 706, a user input unit 707, an interface unit 708, a memory 709, and a processor 710.

[0171] Those skilled in the art can understand that the electronic device 700 may further include a power source (such as a battery) for powering each component. The power source can be logically connected to the processor 710 through a power management system, so as to implement functions such as management of charging, discharging, and power consumption management through the power management system. Figure 15 The structure of the electronic device shown does not limit the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0172] Among them, the processor 710 is used to determine the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device.

[0173] The processor 710 is further used to transfer the load information to the scheduler corresponding to the working scenario for processing.

[0174] The processor 710 is further used to determine the processor parameters corresponding to the working scenario according to the processed load information.

[0175] In the embodiment of the present application, according to the load characteristic information and service quality information of the working scenario of the electronic device, the load information of the working scenario is determined; the load information is transferred to the scheduler corresponding to the working scenario for processing; the processor parameters corresponding to the working scenario are determined according to the processed load information. In the embodiment of the present application, based on the load characteristic information and service quality information of the working scenario of the electronic device, the load information of the working scenario is predicted, and the corresponding scheduler is selected to process the load information, and then the processor parameters required for the working scenario are scheduled based on the processed load information. In this way, the load information is predicted and resource scheduling is performed based on the relevant information of the working scenario, without relying on manual debugging, which improves the intelligence, comprehensiveness, and accuracy of processor scheduling.

[0176] Optionally, the processor 710 is further used to: use a target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device; obtain the load characteristic information of the working scenario.

[0177] In the above embodiments provided by the present application, a target model is used to identify the working scenario of an electronic device according to the interface information and running information of the electronic device; and the load characteristic information of the working scenario is obtained. In this way, by using the target model to identify the working scenario of the electronic device according to the relevant information of the electronic device, it is not necessary to rely on manual debugging, which improves the intelligence of the processor scheduling, reduces the human and material resources consumed by the processor scheduling, and based on the target model, most task scenarios and key task scenarios can be identified, which improves the comprehensiveness and accuracy of the processor scheduling.

[0178] Optionally, the processor 710 is specifically configured to: use the target model to capture the display interface of the electronic device, perform interface recognition on the display interface to obtain an interface recognition result; obtain the running load and running core information of the electronic device; and determine the working scenario of the electronic device according to the interface recognition result, the running load and the running core information.

[0179] In the above embodiments provided by the present application, in the process of using the target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device, the target model is used to capture the display interface of the electronic device, perform interface recognition on the display interface to obtain an interface recognition result; obtain the running load and running core information of the electronic device; and determine the working scenario of the electronic device according to the interface recognition result, the running load and the running core information. In this way, based on the screenshot recognition ability of the target model, combined with the running load and running core information of the electronic device, the working scenario of the electronic device is identified, without relying on manual debugging, which improves the intelligence of the scenario recognition, and based on the target model, most task scenarios and key task scenarios can be identified, which improves the comprehensiveness and accuracy of the scenario recognition.

[0180] Optionally, the processor 710 is specifically configured to: perform interface recognition on the display interface by using the target model to determine the interface elements in the display interface; determine the target area or target element in the display interface according to the position information and semantic information of the interface elements; and determine the interface recognition result according to the target area or target element.

[0181] In the above embodiments provided by the present application, in the process of using the target model to capture the display interface of the electronic device and perform interface recognition on the display interface to obtain an interface recognition result, the target model is used to perform interface recognition on the display interface to determine the interface elements in the display interface; determine the target area or target element in the display interface according to the position information and semantic information of the interface elements; and determine the interface recognition result according to the target area or target element. In this way, by using the screenshot recognition ability of the target model, the working scenario of the electronic device is identified, without relying on manual debugging, which improves the intelligence of the scenario recognition, and most task scenarios and key task scenarios can be identified, which improves the comprehensiveness and accuracy of the scenario recognition.

[0182] Optionally, the processor 710 is specifically configured to: when the working scenario includes a game scenario, transmit the load information of the game scenario to the first scheduler for processing; when the working scenario includes a non-game scenario, transmit the load information of the non-game scenario to the second scheduler for processing; wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario.

[0183] In the above embodiments provided by the present application, during the process of transmitting the load information to the scheduler corresponding to the working scenario for processing, when the working scenario includes a game scenario, the load information of the game scenario is transmitted to the first scheduler for processing; when the working scenario includes a non-game scenario, the load information of the non-game scenario is transmitted to the second scheduler for processing; wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario. In this way, customizing the scheduler for the game scenario can better allocate the processor performance based on the load of the actual task scenario.

[0184] Optionally, the processor 710 is specifically configured to: determine the core configuration information according to the game scenario and the load information; determine the processor characteristic information according to the core configuration information; and determine the interface driver information of the processor according to the processor characteristic information.

[0185] In the above embodiments provided by the present application, during the process of transmitting the load information of the game scenario to the first scheduler for processing, the core configuration information is determined according to the game scenario and the load information; the processor characteristic information is determined according to the core configuration information; and the interface driver information of the processor is determined according to the processor characteristic information. In this way, customizing the scheduler for the game scenario can better allocate the processor performance based on the load of the actual game scenario.

[0186] Optionally, the processor 710 is specifically configured to: use the target model to determine the target cluster processor core and the processor frequency of each processor core in the target cluster processor core according to the processed load information, the processor type, the scenario type, and the performance information of each cluster of processor cores.

[0187] In the above embodiments provided by the present application, during the process of determining the processor parameters corresponding to the working scenario according to the processed load information, the target model is used to determine the target cluster processor core and the processor frequency of each processor core in the target cluster processor core according to the processed load information, the processor type, the scenario type, and the performance information of each cluster of processor cores. In this way, the processor scheduling for the task scenario based on the target model improves the intelligence, comprehensiveness, and accuracy of the processor scheduling.

[0188] Optionally, before using the target model to identify the working scenario of the electronic device based on the interface information and operating information of the electronic device, the processor 710 is further configured to: obtain a first correspondence between the first type of task scenario and the task load and core configuration information; obtain a second correspondence between the second type of task scenario and the task load and core configuration information; obtain a third correspondence between the third type of task scenario and the task load and core configuration information; train the target model according to the first correspondence, the second correspondence, and the third correspondence; and update the trained target model according to the feedback information of the electronic device for each task scenario.

[0189] In the above embodiments provided by the present application, before using the target model to identify the working scenario of the electronic device based on the interface information and operating information of the electronic device, a first correspondence between the first type of task scenario and the task load and core configuration information is obtained; a second correspondence between the second type of task scenario and the task load and core configuration information is obtained; a third correspondence between the third type of task scenario and the task load and core configuration information is obtained; the target model is trained according to the first correspondence, the second correspondence, and the third correspondence; and the trained target model is updated according to the feedback information of the electronic device for each task scenario. In this way, the target model can be endowed with scene perception ability and the processor intelligent scheduling ability, improving the intelligence, comprehensiveness, and accuracy of the processor scheduling.

[0190] Optionally, after determining the processor parameters corresponding to the working scenario according to the processed load information, the processor 710 is further configured to: update the load characteristic information of the working scenario according to the feedback information of the electronic device for processing the working scenario.

[0191] In the above embodiments provided by the present application, after determining the processor parameters corresponding to the working scenario according to the processed load information, the load characteristic information of the working scenario is updated according to the feedback information of the electronic device for processing the working scenario. In this way, the accuracy of subsequent processor scheduling based on the load characteristic information is improved.

[0192] It should be understood that in the embodiments of the present application, the input unit 704 may include a Graphics Processing Unit (GPU) 7041 and a microphone 7042. The GPU 7041 processes the image data of static pictures or videos obtained by an image capturing device (such as a camera) in the video capturing mode or the image capturing mode. The display unit 706 may include a display panel 7061, and the display panel 7061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 707 includes at least one of a touch panel 7071 and other input devices 7072. The touch panel 7071 is also referred to as a touch screen. The touch panel 7071 may include two parts, a touch detection device and a touch controller. The other input devices 7072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, power on / off keys, etc.), a trackball, a mouse, and a joystick, which will not be elaborated here.

[0193] The memory 709 can be used to store software programs and various data. The memory 709 mainly includes a first storage area for storing programs or instructions and a second storage area for storing data. Among them, the first storage area can store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 709 may include a volatile memory or a non-volatile memory, or the memory 709 may include both a volatile memory and a non-volatile memory. Among them, the non-volatile memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash memory. The volatile memory may be a Random Access Memory (RAM), a Static RAM (SRAM), a Dynamic RAM (DRAM), a Synchronous DRAM (SDRAM), a Double Data Rate SDRAM (DDR SDRAM), an Enhanced SDRAM (ESDRAM), a Synch link DRAM (SLDRAM), and a Direct Rambus RAM (DRRAM). The memory 709 in the embodiments of the present application includes, but is not limited to, these and any other suitable types of memories.

[0194] The processor 710 may include one or more processing units; optionally, the processor 710 integrates an application processor and a modem processor. Among them, the application processor mainly processes operations related to the operating system, user interface, application programs, etc., and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above-mentioned modem processor may not be integrated into the processor 710 either.

[0195] The embodiments of the present application also provide a readable storage medium. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, each process of the above-mentioned resource scheduling method embodiments is implemented, and the same technical effects can be achieved. To avoid repetition, details are not described here again.

[0196] Among them, the processor is the processor in the electronic device in the above-mentioned embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disks, or optical discs, etc.

[0197] The embodiments of the present application further provide a chip. The chip includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement each process of the above-mentioned resource scheduling method embodiments, and the same technical effects can be achieved. To avoid repetition, details are not described here again.

[0198] It should be understood that the chip mentioned in the embodiments of the present application may also be referred to as a system-on-chip, system chip, chip system, or system-on-chip, etc.

[0199] The embodiments of the present application provide a computer program product. The program product is stored in a storage medium. The program product is executed by at least one processor to implement each process of the resource scheduling method embodiments as described above, and the same technical effects can be achieved. To avoid repetition, details are not described here again.

[0200] It should be noted that in this article, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without more limitations, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, the features described with reference to certain examples may be combined in other examples.

[0201] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to enable a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present application.

[0202] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them fall within the protection scope of the present application.

Claims

1. A resource scheduling method, characterized in that, Including: Determine the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device; Transmit the load information to the scheduler corresponding to the working scenario for processing; Determine the processor parameters corresponding to the working scenario according to the processed load information.

2. The resource scheduling method according to claim 1, wherein Before determining the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device, the resource scheduling method further includes: Use the target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device; Obtain the load characteristic information of the working scenario.

3. The resource scheduling method according to claim 2, wherein The using the target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device includes: Use the target model to intercept the display interface of the electronic device, and perform interface recognition on the display interface to obtain an interface recognition result; Obtain the running load and running core information of the electronic device; Determine the working scenario of the electronic device according to the interface recognition result, the running load, and the running core information.

4. The resource scheduling method according to claim 1, wherein The transmitting the load information to the scheduler corresponding to the working scenario for processing includes: In the case where the working scenario includes a game scenario, transmit the load information of the game scenario to the first scheduler for processing; In the case where the working scenario includes a non-game scenario, transmit the load information of the non-game scenario to the second scheduler for processing; Wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario.

5. The resource scheduling method according to claim 4, wherein The transmitting the load information of the game scenario to the first scheduler for processing includes: Determine the core configuration information according to the game scenario and the load information; Determine the processor characteristic information according to the core configuration information; Determine the interface driver information of the processor according to the processor characteristic information.

6. The resource scheduling method according to claim 2, wherein The processor parameters include the processor core and the processor frequency. The determining the processor parameters corresponding to the working scenario according to the processed load information includes: Use the target model to determine the target cluster processor core and the processor frequency of each processor core in the target cluster processor core according to the processed load information, the processor type, the scenario type, and the performance information of each cluster of processor cores.

7. The resource scheduling method according to claim 2, wherein Before using the target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device, the resource scheduling method further includes: Obtain the first correspondence between the first type of task scenario and the task load and core configuration information; Obtain the second correspondence between the second type of task scenario and the task load and core configuration information; Obtain the third correspondence between the third type of task scenario and the task load and core configuration information; Train the target model according to the first correspondence, the second correspondence, and the third correspondence; Update the trained target model according to the feedback information of the electronic device for each task scenario.

8. A resource scheduling system, characterized in that, Including: The kernel layer includes a load decision module, and the load decision module is used to determine the load information of the working scenario according to the load characteristic information and service quality information of the working scenario of the electronic device; The load decision module is further used to transfer the load information to the scheduler corresponding to the working scenario for processing; The hardware layer is used to determine the processor parameters corresponding to the working scenario according to the processed load information.

9. The resource scheduling system according to claim 8, wherein It further includes: The application layer includes a scenario perception engine module, and the scenario perception engine module is used to use a target model to identify the working scenario of the electronic device according to the interface information and running information of the electronic device; The scenario perception engine module is further used to obtain the load characteristic information of the working scenario.

10. The resource scheduling system according to claim 9, wherein The scenario perception engine module specifically is used for: Using the target model, intercepting the display interface of the electronic device, and performing interface recognition on the display interface to obtain an interface recognition result; Obtaining the running load and running core information of the electronic device; Determining the working scenario of the electronic device according to the interface recognition result, the running load, and the running core information.

11. The resource scheduling system according to claim 8, wherein The load decision module specifically is used for: In the case where the working scenario includes a game scenario, transferring the load information of the game scenario to the first scheduler for processing; In the case where the working scenario includes a non-game scenario, transferring the load information of the non-game scenario to the second scheduler for processing; Wherein, the first scheduler is an extensible scheduler, and the scheduling policy of the first scheduler is determined according to the load characteristic information of the game scenario.

12. The resource scheduling system according to claim 11, wherein The load decision module specifically is used for: Determining core configuration information according to the game scenario and the load information; Determining processor characteristic information according to the core configuration information; Determining the interface driver information of the processor according to the processor characteristic information.

13. The resource scheduling system according to claim 9, wherein The processor parameters include a processor core and a processor frequency, and the hardware layer specifically is used for: Using the target model, determining the target cluster processor core and the processor frequency of each processor core in the target cluster processor core according to the processed load information, processor type, scenario type, and performance information of each cluster of processor cores.

14. The resource scheduling system according to claim 9, wherein The scenario perception engine module is further used for: Obtaining a first correspondence between a first type of task scenario and task load and core configuration information; Obtaining a second correspondence between a second type of task scenario and task load and core configuration information; Obtaining a third correspondence between a third type of task scenario and task load and core configuration information; Training the target model according to the first correspondence, the second correspondence, and the third correspondence; Updating the trained target model according to the feedback information of the electronic device for each task scenario.

15. An electronic device, characterized in that, It includes a processor and a memory, and the memory stores a program or instruction that can run on the processor. When the program or instruction is executed by the processor, the steps of the resource scheduling method according to any one of claims 1 to 7 are implemented.

Images