Energy efficiency optimization method suitable for mobile application processor

Abstract

The embodiment of the invention provides an energy efficiency optimization method suitable for a mobile application processor, and the method comprises the steps of collecting target environment state data of system software and hardware during the operation of a mobile game, and constructing an FPS-Meet scene data set and an FPS-Loss data set according to whether frame loss occurs or not; obtaining an FPS-Meet scene frequency modulation model and an FPS-Loss scene frequency modulation model through the data set; writing the FPS-Meet scene frequency modulation model and the FPS-Loss scene frequency modulation model into an operating system kernel; when the mobile application runs, reading the software and hardware environment state of the system in the kernel of the operating system at regular time, and calling the corresponding model according to whether the current game loses frames. According to the method, scene-adaptive processor frequency adjustment according to the scene of the mobile application can be realized, the required frequency can be accurately calculated when the load fluctuation is calculated, and extra energy consumption overhead caused by performance excess can be avoided; and when the calculation load fluctuates violently, the frequency is increased in time to prevent frame loss and lagging from influencing the user experience.

Description

technical field [0001] The present application relates to the field of dynamic frequency adjustment of computer operating systems, and in particular, relates to an energy efficiency optimization method suitable for mobile application processors. Background technique [0002] With the development of mobile processors, mobile game applications are becoming more and more complex, dazzling rendering effects, rich game scenes, and even professional game engine support. Accompanied by the above characteristics, mobile games require a powerful mobile CPU to handle complex game scene processing tasks and graphics rendering tasks. The mobile processor adjusts performance according to actual needs (through frequency modulation). Insufficient adjusted performance will cause frame loss and freeze, and excessive adjusted performance will cause heat generation and high energy consumption. Both of which will seriously affect user experience. Therefore, the adjustment strategy is the key. ...

AI Technical Summary

Problems solved by technology

[0004] However, due to its implementation mechanism (mainly based on the characteristics of desktop applications), the optimization effect for mobile applications is extremely limited, especially for mobile applications that require high real-time and interactivity, and the shortcomings are very prominent:

[0005] (1) The frequency modulation accuracy is insufficient, and the adjustable frequency is divided into several different levels, which cannot meet the complex and diverse computing performance requirements of mobile games;

[0006] (2) Focusing on ensuring system stability, the idea of ​​frequency modulation is conservative, and the balance between performance and power consumption is insufficiently considered. In most cases, there is excess performance, which brings additional energy consumption;

[0007] (3) Predicting frequency changes based on historical load sampling has high latency and cannot meet the real-time requirements of frequency modulation. For sudden high-performance scenarios, etc., the up-frequency response cannot be made in time, resulting in game frame loss and freezing, affecting user experience.

[0008] In related technologies, high-performance, high-interaction applications based on mobile terminals can achieve precise frequency modulation, avoid frame loss and freeze, and improve energy efficiency. At present, there is no effective solution

Images