JP6789861B2 - Operating frequency control device, operating frequency control program, and operating frequency control method - Google Patents

Description

This case relates to an operating frequency control device, an operating frequency control program, and an operating frequency control method.

A technique for controlling the operating frequency of a Central Processing Unit (CPU) in a mobile information terminal such as a smartphone is known (see, for example, Patent Document 1). Further, there is also known a control method for setting an operating frequency in response to a state change of a CPU processing load (hereinafter, simply referred to as a load). The control method may be referred to as governor control, for example (see, for example, Patent Document 2).

As a result, the application software installed on the mobile information terminal (hereinafter, simply referred to as an application) is executed, and even if the CPU load state changes variously according to the execution of the application, the operation according to the application is controlled by the governor. The CPU operates at the frequency. As a result, the performance of the mobile information terminal is ensured, and the generation of unnecessary power consumption is suppressed.

Japanese Unexamined Patent Publication No. 2014-182736 Japanese Unexamined Patent Publication No. 2015-043165

By the way, the above-mentioned apps are roughly divided into apps installed in advance (for example, before shipping) on mobile information terminals (hereinafter referred to as initial apps) and apps installed later (for example, after shipping) (hereinafter referred to as additional apps). Will be done. Here, in the case of the initial application, the CPU performance required by the initial application can be confirmed in advance, so that a fixed operating frequency that realizes the performance can be individually prepared by design.

However, in the case of the additional application, the CPU performance required by the additional application cannot be confirmed in advance. Therefore, it is not possible to individually prepare a fixed operating frequency that realizes the performance by design. For example, when an additional application is executed based on the operating frequency prepared for the initial application, the CPU performance may be insufficient, or conversely, the CPU performance may be excessive. If the performance of the CPU is insufficient, the performance of the mobile information terminal is lowered, and conversely, if the performance of the CPU is excessive, there is a problem that wasteful power consumption is generated.

Therefore, in one aspect, it is an object of the present invention to provide an operating frequency control device, an operating frequency control program, and an operating frequency control method capable of adjusting the performance of the CPU with respect to an additional application.

In one embodiment, the operating frequency control device causes an additional application to be operated based on the history of the operation when the statistical information of the task generated based on the operation satisfies a predetermined condition. An operating frequency control device including a processing unit that executes processing, which calculates parameters for controlling the operating frequency of the CPU based on the load of the CPU during execution of the additional application.

You can adjust the CPU performance for additional apps.

FIG. 1 is an example of a mobile information terminal. FIG. 2 is an example of the hardware configuration of the mobile information terminal. FIG. 3 is an example of a functional block diagram of a mobile information terminal. FIG. 4 is an example of characteristic information. FIG. 5 is an example of a target load table. FIG. 6 is a part (No. 1) of the operating frequencies before and after the approximation, which changes with the passage of time. FIG. 7 is a part (No. 2) of the operating frequencies before and after the approximation, which changes with the passage of time. FIG. 8 is a part (No. 3) of the operating frequencies before and after the approximation, which changes with the passage of time. FIG. 9 is an example of the operation of the operating frequency control device. FIG. 10 is a flowchart showing an example of the event history holding process. FIG. 11 is a flowchart showing an example of the parameter calculation process.

Hereinafter, a mode for carrying out this case will be described with reference to the drawings.

FIG. 1 is an example of a mobile information terminal 10. The mobile information terminal 10 includes a smart device powered by a battery, such as a tablet terminal or a smartphone. The mobile information terminal 10 is not limited to a smart device, and may be a wearable terminal such as a smart watch. As shown in FIG. 1, the mobile information terminal 10 includes an input unit 11. Although the details will be described later, when the input unit 11 is operated by the user, the input unit 11 accepts the input according to the operation content as the user operation.

FIG. 2 is an example of the hardware configuration of the mobile information terminal 10.

As shown in FIG. 2, the mobile information terminal 10 includes a Central Processing Unit (CPU) 100A, a Random Access Memory (RAM) 100B, a Read Only Memory (ROM) 100C, a Non-Volatile Memory (NVM) 100D, and a Radio Frequency (RF). ) Includes circuit 100E. An antenna 100E'is connected to the RF circuit 100E. A CPU that realizes a communication function may be used instead of the RF circuit 100E.

Further, the personal digital assistant 10 includes a Graphics Processing Unit (GPU) 100F, a camera 100G, a touch panel 100H as an input unit 11, a display 100I, and a microphone and a speaker (hereinafter referred to as a microphone / speaker) 100J. The CPU 100A to the microphone / speaker 100J are connected to each other by the internal bus 100K. A computer is realized by at least the CPU 100A and the RAM 100B working together.

In the RAM 100B described above, the program stored in the ROM 100C or the NVM 100D is stored by the CPU 100A. When the CPU 100A executes the stored program, various processes described later are executed. The program may be adapted to the flowchart described later.

FIG. 3 is an example of a functional block diagram of the mobile information terminal 10.

As shown in FIG. 3, the mobile information terminal 10 includes an operating frequency control device 100. In particular, the operating frequency control device 100 includes an event control unit 101, a power consumption measurement unit 102, and an event history management unit 103 as processing units. Further, the operating frequency control device 100 includes a characteristic information storage unit 104, an application execution unit 105, and a load history management unit 106 as processing units. Further, the operating frequency control device 100 includes a parameter calculation unit 107, a parameter conversion unit 108, an operating frequency control unit 109, and a parameter storage unit 110 as processing units.

Since the CPU 100A and the touch panel 100H are hardware, they are shown by broken lines. Here, the characteristic information storage unit 104 is realized by, for example, the ROM 100C or the NVM 100D described above. The parameter storage unit 110 is realized by, for example, the above-mentioned RAM 100B or NVM 100D. Each processing unit (for example, event control unit 101) excluding the characteristic information storage unit 104 and the parameter storage unit 110 is realized by the CPU 100A described above.

The event control unit 101 receives the user operation output from the touch panel 100H. In particular, the event control unit 101 accepts user operations in the normal mode described later. User operations include, for example, screen operations and operations for starting additional applications. The normal mode refers to an operation mode in which the operation frequency of the CPU 100A is changed according to the load of the CPU 100A (specifically, the CPU usage rate).

When the event control unit 101 accepts a user operation, it generates event information according to the user operation and at least one task (hereinafter referred to as a task) to be executed by the CPU 100A according to the user operation, and generates at least one event information and the generated event information. It is output to the event history management unit 103 in association with the task. Therefore, for example, when the event control unit 101 receives a user operation for invoking the additional application, the event control unit 101 generates at least one task required for executing the additional application, and outputs the generated task to the CPU 100A. The CPU 100A executes an additional application by sequentially processing tasks.

The power consumption measuring unit 102 measures the power consumption of the CPU 100A. More specifically, the power consumption measuring unit 102 measures the power consumption of the task processed by the CPU 100A. When the power consumption measuring unit 102 measures the power consumption of the task, the power consumption measuring unit 102 outputs the measured power consumption of the task to the event history management unit 103.

The event history management unit 103 manages the history of event information. Specifically, the event history management unit 103 holds the event information output from the event control unit 101 for each task. As a result, for example, the event history management unit 103 holds event information according to the user operation for starting the additional application having a high start load. Similarly, for example, the event history management unit 103 holds event information according to a user operation up to a screen (specifically, a frame of a camera 100G or the like) in which a high load state continues. For example, the event history management unit 103 holds event information according to user operations up to a process that is very frequently used. The event information held by the event history management unit 103 is used in the measurement mode described later.

In addition, the event history management unit 103 holds the event information in association with the task, and also holds the power consumption of each task output from the power consumption measurement unit 102 in association with each other. Therefore, the event history management unit 103 can generate statistical information about the task. More specifically, the event history management unit 103 can generate a first statistical information indicating the frequency of use of the task and a second statistical information representing the average power consumption of the task. In particular, the event history management unit 103 can generate the second statistical information based on the power consumption of the task output from the power consumption measurement unit 102.

Then, when the statistical information satisfies the predetermined condition, the event history management unit 103 sets itself in the measurement mode, and at the timing when the user operation is not performed, the application execution unit 105 and the parameter calculation unit 107 are switched to the measurement mode. Instruct the migration of. For example, the event history management unit 103 instructs the application execution unit 105 and the parameter calculation unit 107 to shift to the measurement mode when both the first statistical information and the second statistical information exceed the threshold values specified in advance. To do.

The measurement mode is an operation mode for calculating governor parameters used in the additional application. The governor parameter is a parameter that determines the characteristics of governor control (for example, emphasis on operability, emphasis on power consumption, balance between operability and power consumption, etc.). Details will be described later, but for example, above_hispeed_delay and min_sample_time are known as governor parameters.

The characteristic information storage unit 104 stores characteristic information representing the characteristics of the CPU 100A. The characteristic information is measured in advance, and the characteristic information storage unit 104 stores the characteristic information measured in advance. Specifically, as shown in FIG. 4, the characteristic information storage unit 104 stores the power consumption when the CPU 100A processes a task of the same load for each operating frequency that can be set in the CPU 100A as characteristic information. In particular, the characteristic information storage unit 104 stores characteristic information arranged in ascending order of power consumption.

When the event history management unit 103 instructs the application execution unit 105 to shift to the measurement mode, the application execution unit 105 acquires event information from the event history management unit 103 and causes the event control unit 101 to start an additional application according to the event information. As a result, the event control unit 101 generates at least one task required for executing the additional application, and outputs the generated task to the CPU 100A. Therefore, the CPU 100A executes the additional application again.

The load history management unit 106 manages the load history of the CPU 100A. Specifically, when the CPU 100A starts executing the additional application, the load history management unit 106 measures the load of the CPU 100A during the execution of the additional application, and holds the measured load as an actually measured load.

When the event history management unit 103 instructs the parameter calculation unit 107 to shift to the measurement mode, the parameter conversion unit 108 invalidates the function in the normal mode and enables the function in the measurement mode. When the parameter conversion unit 108 activates the function in the measurement mode, the operating frequency control unit 109 operates the CPU 100A at the maximum frequency. Therefore, the CPU 100A re-executes the additional application at the maximum frequency.

Further, the parameter calculation unit 107 acquires the actually measured load from the load history management unit 106. Since the CPU 100A executes the additional application at the maximum frequency, the parameter calculation unit 107 acquires the actually measured load of the CPU 100A operating at the maximum frequency. When the parameter calculation unit 107 acquires the actually measured load, it calculates the provisional frequency by using the target load, the maximum frequency of the CPU 100A, and the following calculation formula (1) or (2).
Temporary frequency = maximum frequency x measured load ÷ target load (measured load ≤ target load) ... (1)
Temporary frequency = maximum frequency (measured load> target load) ... (2)

In particular, since the parameter calculation unit 107 constantly or periodically calculates the provisional frequency while the CPU 100A is executing the additional application, the parameter calculation unit 107 obtains the provisional frequency that changes with the passage of time. Can be done.

Here, the target load means a guideline of the load targeted by the CPU 100A when the CPU 100A operates in the measurement mode. The target load may be a fixed value that is not related to the actually measured load, or may be a variable value that is dynamically variable according to the actually measured load. In the former case, the provisional frequency can be calculated quickly, and in the latter case, waste of power consumption can be eliminated.

Specifically, as shown in FIG. 5, the parameter calculation unit 107 holds a target load table in which lower and upper limits are predetermined for each measured load. Then, when the actually measured load is lower than the first predetermined value (for example, 20%), the parameter calculation unit 107 uses the target load (for example, 50%) that greatly deviates from the actually measured load to the high load side. According to the calculation formula (1), the provisional frequency is lowered as a result. On the contrary, when the actually measured load is higher than the second predetermined value (for example, 80%), the parameter calculation unit 107 uses a target load (for example, 90%) that does not deviate so much from the actually measured load.

According to the calculation formula (1), the provisional frequency is eventually close to the maximum frequency, and according to the calculation formula (2), the provisional frequency is eventually the same as the maximum frequency. When the actually measured load is included between the first predetermined value and the second predetermined value, the target loads of 70% and 90% are used in FIG. 5, and these target loads are the characteristic information storage unit. It is calculated in advance using the characteristic information stored in 104. Specifically, the target load that approaches the most efficient operating frequency in terms of power consumption is calculated in advance based on the characteristic information.

Here, the provisional frequency calculated by the parameter calculation unit 107 is not necessarily the operating frequency that can be set in the CPU 100A (specifically, each operating frequency shown in FIG. 4). Therefore, the parameter calculation unit 107 replaces the provisional frequency with an operating frequency that can be set in the CPU 100A. Specifically, the parameter calculation unit 107 replaces the provisional frequency with the operating frequency according to any of the following four predetermined patterns (A) to (D).

(A) Of the two operating frequencies sandwiching the calculated provisional frequency, the more efficient operating frequency is selected as the characteristic of the CPU 100A based on the characteristic information, and the provisional frequency is replaced with the selected operating frequency.
(B) Of the two operating frequencies that sandwich the calculated provisional frequency, the operating frequency with the smaller difference from the provisional frequency is selected, and the provisional frequency is replaced with the selected operating frequency.
(C) Of the two operating frequencies that sandwich the calculated provisional frequency, the operating frequency that is smaller than the provisional frequency is always selected, and the provisional frequency is replaced with the selected operating frequency.
(D) Of the two operating frequencies that sandwich the calculated provisional frequency, the operating frequency that is always larger than the provisional frequency is always selected, and the provisional frequency is replaced with the selected operating frequency.

As a result, the provisional frequency that changes with the passage of time is replaced with the operating frequency that can be set in the CPU 100A. However, the operating frequency changes very finely over time. Therefore, in order to reduce the frequency of the overall change, the parameter calculation unit 107 approximates the change in the operating frequency based on any one of the plurality of governor parameters stored in advance by the parameter storage unit 110 or a combination thereof.

Here, the approximation of the change in the operating frequency will be specifically described with reference to FIGS. 6 to 8.

6 to 8 are a part of the operating frequencies before and after the approximation (No. 1 to No. 3) which change with the passage of time. In other words, by combining a part of the operating frequencies before approximation shown in FIGS. 6 to 8, the entire operating frequency that can be set in the CPU 100A, which changes with the passage of time, is completed.

When the parameter calculation unit 107 approximates the change in the operating frequency, first, the parameter calculation unit 107 has a specific threshold value for the duration of a constant operating frequency from the operating frequencies before approximation shown in the upper part of FIGS. 6 to 8. Identify the part that is less than or equal to the time. Specifically, if the operating frequency is the operating frequency before approximation shown in the upper part of FIG. 6, the parameter calculation unit 107 specifies the portion A. Similarly, if the operating frequency is the operating frequency before approximation shown in the upper part of FIG. 7, the parameter calculation unit 107 specifies the portion B. If it is the operating frequency before approximation shown in the upper part of FIG. 8, the parameter calculation unit 107 specifies the part C.

Next, the parameter calculation unit 107 replaces the operating frequency of the specified portion with a constant operating frequency that exists immediately before the specified portion and continues for a longer time than the portion, and replaces the replaced portion with the interval of the immediately preceding operating frequency. Join. Specifically, as shown in the lower part of FIG. 6, the parameter calculation unit 107 replaces the operating frequency of the portion A with the operating frequency that exists immediately before, and replaces the portion A and the section A'of the operating frequency that exists immediately before. Join. Similarly, as shown in the lower part of FIG. 7, the parameter calculation unit 107 replaces the operating frequency of the portion B with the operating frequency that exists immediately before, and combines the operating frequency section B'of the operating frequency that exists immediately before the portion B. .. Further, as shown in the lower part of FIG. 8, the parameter calculation unit 107 replaces the operating frequency of the partial C with the operating frequency existing immediately before, and combines the partial C with the section C'of the operating frequency existing immediately before. .. As a result, small changes disappear, and an operating frequency with a large change and a constant operating frequency that continues for a long time remains.

Then, the parameter calculation unit 107 determines the governor parameter based on the remaining operating frequency. Specifically, regarding the governor parameter (specifically, hispeed_freq) indicating the next operating frequency to be taken when the frequency is at the lowest frequency and the governor parameter (specifically, above_hispeed_delay), the parameter calculation unit 107 has each Based on the statistical values of, each mode value is determined as a governor parameter.

Further, the parameter calculation unit 107 determines governor parameters (specifically, target_load and hispeed_load) related to the load based on the target load. Further, regarding the governor parameter (specifically, min_sample_time) representing the duration after the operating frequency is set for each operating frequency, the parameter calculation unit 107 determines the minimum value as the governor parameter based on the statistical value. Regarding the governor parameter (specifically, timer_slack) representing the frequency maintenance time from the start of idle, the parameter calculation unit 107 determines a fixed value of the initial setting as the governor parameter.

When the parameter calculation unit 107 determines the governor parameter, the application execution unit 105 causes the event control unit 101 to start the additional application. Then, the parameter calculation unit 107 is the power consumption measured by the power consumption measurement unit 102 when the additional application is executed based on the default governor parameter, and the additional application is executed based on the determined governor parameter. The power consumption measured by the power consumption measuring unit 102 is compared. As a result of comparison, when the determined governor parameter consumes less power than the default governor parameter, the parameter calculation unit 107 stores the determined governor parameter in the parameter storage unit 110 in association with the information of the additional application.

The parameter conversion unit 108 converts the governor parameter into an operating frequency. More specifically, when the additional application is executed, the parameter conversion unit 108 acquires the governor parameter corresponding to the additional application from the governor parameters stored in the parameter storage unit 110. When the parameter conversion unit 108 acquires the governor parameter corresponding to the additional application, the parameter conversion unit 108 converts the acquired governor parameter into an operating frequency and outputs the acquired governor parameter to the operating frequency control unit 109.

The operating frequency control unit 109 operates the CPU 100A at the operating frequency output from the parameter conversion unit 108. Therefore, the CPU 100A operates according to the operating frequency output from the parameter conversion unit 108.

Next, the operation of the operating frequency control device 100 will be described with reference to FIG.

FIG. 9 is an example of the operation of the operating frequency control device 100. As shown in FIG. 9, the event history management unit 103 first executes the event history holding process (step S101). Although the details will be described later, the event history holding process is a process of holding the event information output from the event control unit 101 in association with the task. When the event history holding process is completed, the event history management unit 103 then determines whether or not the task statistical information satisfies a predetermined condition (step S102).

When the statistical information of the task does not satisfy the predetermined condition (step S102: NO), the event history management unit 103 re-executes the process of step S101. On the other hand, when the statistical information of the task satisfies a predetermined condition (step S102: YES), the parameter calculation unit 107 executes the parameter calculation process (step S103). Although the details will be described later, the parameter calculation process is a process of calculating the governor parameter that controls the operating frequency of the CPU 100A based on the load of the CPU 100A during the execution of the additional application. When the parameter calculation process is completed, the parameter calculation unit 107 ends the process.

Next, the details of the above-mentioned event history holding process will be described with reference to FIG.

FIG. 10 is a flowchart showing an example of the event history holding process. For example, when the power switch of the mobile information terminal 10 is switched from off to on, the event history management unit 103 sets itself to the normal mode (step S201). When the process of step S201 is completed and the event control unit 101 receives a user operation for invoking an additional application from the touch panel 100H, the event control unit 101 generates event information (step S202), and the event history management unit 103 controls the event. The event information generated by the unit 101 is held (step S203).

Further, when the event control unit 101 receives the user operation for starting the additional application, the additional application is started (step S204), and when the additional application is started, the CPU 100A executes the additional application. The order of the processing in steps S202 and S203 and the processing in step S204 may be reversed. When the process of step S204 is completed, the event history management unit 103 then acquires information (for example, application name, screen name, etc.) related to various tasks generated in response to the user operation from the operating system (OS) (step S205). ). When the event history management unit 103 acquires the information about the task, it generates the statistical information of the task (step S206). When the process of step S206 is completed, the event history management unit 103 holds the event information and the statistical information in association with each other (step S207). As a result, the event history management unit 103 finishes the event history holding process.

Next, the details of the parameter calculation process will be described with reference to FIG.

FIG. 11 is a flowchart showing an example of the parameter calculation process. When the statistical information generated by the event history management unit 103 satisfies a predetermined condition, the event history management unit 103 sets itself in the measurement mode (step S301). When the event history management unit 103 sets itself to the measurement mode, the parameter calculation unit 107 fixes the operating frequency of the CPU 100A to the maximum frequency (step S302). More specifically, the parameter calculation unit 107 invalidates the function of the parameter conversion unit 108 in the normal mode and enables the function in the measurement mode. As a result, the parameter conversion unit 108 stops the function of outputting the operating frequency corresponding to the load of the CPU 100A to the operating frequency control unit 109, and realizes the function of outputting the maximum frequency to the operating frequency control unit 109. As a result, the operating frequency control unit 109 operates the CPU 100A at the maximum frequency.

When the process of step S302 is completed, the parameter calculation unit 107 then acquires the characteristic information (step S303), and the application execution unit 105 causes the event control unit 101 to start the additional application (step S304). More specifically, when the event history management unit 103 instructs the application execution unit 105 to shift to the measurement mode, the application execution unit 105 causes the event control unit 101 to start an additional application according to the event information. As a result, the CPU 100A executes the additional application again.

When the process of step S304 is completed, the load history management unit 106 measures the load of the CPU 100A during execution of the additional application (step S305). When the process of step S305 is completed, the parameter calculation unit 107 then converts the actually measured load measured by the load history management unit 106 into a provisional frequency (step S306). More specifically, the parameter calculation unit 107 uses the measured load, the target load determined according to the measured load, the maximum frequency of the CPU 100A, and the predetermined calculation formula (1) or (2) described above to perform the measured load. Convert to a provisional frequency.

When the process of step S306 is completed, the parameter calculation unit 107 then calculates the governor parameter (step S307). More specifically, the parameter calculation unit 107 replaces the provisional frequency with an operating frequency that can be set in the CPU 100A, approximates the operating frequency, and determines various governor parameters using the approximated operating frequency.

When the process of step S307 is completed, the power consumption measuring unit 102 measures the power consumption (step S308). More specifically, the power consumption measuring unit 102 measures the power consumption of the CPU 100A that operates based on the default governor parameter and the power consumption of the CPU 100A that operates based on the determined governor parameter. When the process of step S308 is completed, the parameter calculation unit 107 stores the governor parameter whose minimum power consumption is measured from the power consumption measured by the power consumption measurement unit 102 in the parameter storage unit 110 in association with the information of the additional application. (Step S309). As a result, each time the additional application is executed, the governor parameter corresponding to the additional application is used to operate the CPU 100A.

As described above, according to the present embodiment, the operating frequency control device 100 includes the application execution unit 105 and the parameter calculation unit 107. When the application execution unit 105 determines that the statistical information of the task generated based on the user operation satisfies a predetermined condition specified in advance, the application execution unit 105 is based on the history of the event information held by the event history management unit 103. Start the additional application and let the CPU 100A execute the additional application. The parameter calculation unit 107 calculates the governor parameter based on the measured load of the CPU 100A during the execution of the additional application. Thereby, the performance of the CPU 100A with respect to the additional application can be adjusted. Therefore, the performance (specifically, the response speed or operability) of the mobile information terminal 10 is ensured for the additional application, and the generation of unnecessary power consumption is suppressed. As a result, the rate of decrease in battery capacity is reduced, and the operating time of the portable information terminal 10 is extended.

Although the preferred embodiment of the present invention has been described in detail above, it is not limited to the specific embodiment of the present invention, and various modifications are made within the scope of the gist of the present invention described in the claims.・ Can be changed.

For example, the transition to the measurement mode may be performed based on the user operation, and the activation of the additional application after the transition to the measurement mode may be performed based on the user operation. In this case, the event history management unit 103 does not need to dynamically set itself to the measurement mode, and the application execution unit 105 does not need to dynamically start the additional application in the event control unit 101.

Further, for the replacement of the provisional frequency with the operating frequency, for example, a selection screen may be prepared so that the user can select any of the patterns (A) to (D), or all the patterns (A) to (D). ) May be used to replace the provisional frequency with the operating frequency so that the patterns (A) to (D) for measuring the minimum power consumption are selected. Further, instead of using both the first statistical information and the second statistical information, one of them may be used.

The following additional notes will be further disclosed with respect to the above description.
(Appendix 1) When the statistical information of the task generated based on the operation satisfies the condition specified in advance, the additional application to be operated is executed based on the history of the operation, and the CPU during the execution of the additional application is executed. An operating frequency control device including a processing unit that executes processing, which calculates parameters for controlling the operating frequency of the CPU based on the load of the CPU.
(Appendix 2) When the above conditions are satisfied, the processing unit fixes the operating frequency of the CPU to the maximum frequency, and sets the maximum frequency based on the relationship between the measured load of the CPU and the target load of the CPU. The operating frequency control device according to Appendix 1, wherein the frequency is converted to a provisional frequency and the parameters corresponding to the provisional frequency are calculated.
(Appendix 3) The processing unit measures the power consumption based on the execution of the additional application according to the parameters preset in the mobile information terminal and the calculated parameters, and sets the parameter having the smaller power consumption as the parameter of the additional application. The operating frequency control device according to Appendix 1 or 2, characterized in that it is retained in association with information.
(Supplementary Note 4) The operating frequency control device according to any one of Supplementary note 1 to 3, wherein the statistical information is at least one of the frequency of use and power consumption of the task.
(Appendix 5) When the statistical information of the task generated based on the operation satisfies the condition specified in advance, the additional application to be operated is executed based on the history of the operation, and the CPU during the execution of the additional application is executed. An operating frequency control program that causes a computer to execute a process that calculates parameters that control the operating frequency of the CPU based on the load of the CPU.
(Appendix 6) When the statistical information of the task generated based on the operation satisfies the condition specified in advance, the additional application to be operated is executed based on the history of the operation, and the CPU during the execution of the additional application is executed. A method of operating frequency control in which a computer executes a process for calculating parameters for controlling the operating frequency of the CPU based on the load of the CPU.
(Appendix 7) When the above conditions are satisfied, the operating frequency of the CPU is fixed to the maximum frequency, and the maximum frequency is set to a provisional frequency based on the relationship between the measured load of the CPU and the target load of the CPU. The operating frequency control method according to Appendix 6, wherein the parameter is converted and the parameter is calculated according to the provisional frequency.
(Appendix 8) Measure the power consumption based on the execution of the additional application by each of the parameters preset in the mobile information terminal and the calculated parameters, and save the parameter with the smaller power consumption in association with the information of the additional application. The operating frequency control method according to Appendix 6 or 7, wherein the operating frequency is controlled.
(Supplementary note 9) The operating frequency control method according to any one of Supplementary notes 6 to 8, wherein the statistical information is at least one of the frequency of use and power consumption of the task.

10 Mobile information terminal 100 Operating frequency control device 101 Event control unit 102 Power consumption measurement unit 103 Event history management unit 104 Characteristic information storage unit 105 Application execution unit 106 Load history management unit 107 Parameter calculation unit 108 Parameter conversion unit 109 Operating frequency control unit 110 Parameter storage

Claims (6)

When the statistical information of the task generated based on the operation satisfies the condition specified in advance, the additional application to be operated is executed based on the history of the operation.
A parameter for controlling the operating frequency of the CPU is calculated based on the load of the CPU during execution of the additional application.
An operating frequency control device including a processing unit that executes processing. When the above conditions are satisfied, the processing unit fixes the operating frequency of the CPU to the maximum frequency, and sets the maximum frequency as a provisional frequency based on the relationship between the measured load of the CPU and the target load of the CPU. And calculate the parameters according to the provisional frequency.
The operating frequency control device according to claim 1. The processing unit measures the power consumption based on the execution of the additional application by each of the parameters preset in the mobile information terminal and the calculated parameters, and associates the parameter with the smaller power consumption with the information of the additional application. Hold,
The operating frequency control device according to claim 1 or 2. The statistical information is at least one of the frequency of use and the power consumption of the task.
The operating frequency control device according to any one of claims 1 to 3, wherein the operating frequency control device is characterized. When the statistical information of the task generated based on the operation satisfies the condition specified in advance, the additional application to be operated is executed based on the history of the operation.
A parameter for controlling the operating frequency of the CPU is calculated based on the load of the CPU during execution of the additional application.
An operating frequency control program that causes a computer to perform processing. When the statistical information of the task generated based on the operation satisfies the condition specified in advance, the additional application to be operated is executed based on the history of the operation.
A parameter for controlling the operating frequency of the CPU is calculated based on the load of the CPU during execution of the additional application.
An operating frequency control method in which a computer performs processing.

Images