TWI462020B - Method for managing threads and electronic device using the same method - Google Patents

Description

Thread management method and electronic device thereof

The present invention relates to a thread management mechanism, and in particular to a thread management method and an electronic device thereof implemented by using an execution time scheduling technique.

Currently, applications on electronic devices tend to be diversified and there are threads that typically need to operate multiple segments simultaneously. When the various threads that operate on their own need to operate, the operating system wakes up the processor to execute one or more threads to be processed, so the processor's usage period may be too dense, and the processor's sleep time is reduced. In turn, it consumes too much power.

1A to 1C are schematic diagrams showing the execution timing of a thread and a processor. Referring to FIG. 1A, the horizontal axis is the time axis and the time is increased to the right, and the time points A1 to A6 on the time axis are the execution time that the thread A needs to be executed. The execution of the thread A can be observed from FIG. 1A. The period is, for example, 100 ms. Referring to FIG. 1B, the time points B1 B B3 on the time axis are the execution time that the thread B needs to be executed. The execution period of the thread B can be observed from FIG. 1B, for example, 200 ms.

Thread A and Thread B are threads that operate independently. When both thread A and thread B need to be executed, the processor will wake up to execute thread A and thread B at the appropriate point in time. FIG. 1C is a schematic diagram showing the execution timing when the thread A and the thread B are not properly scheduled by the schedule. It can be observed from FIG. 1C that the processor needs to be every 50 ms. Handling thread A or thread B, the frequency of operation is too high and there is a problem that the peak execution period is too dense, and the processor does not have enough time to enter sleep mode, resulting in higher power consumption. Therefore, how to efficiently schedule multiple threads to improve the power consumption of the processor is still a problem to be solved in this field.

The present invention provides a thread management method and an electronic device thereof, which can plan the execution time of multiple threads in the scheduling memory through a thread manager in the operating system to reduce the frequency of use of the processor, thereby reducing The necessary power consumption.

This case proposes a method of thread management. According to an embodiment of the present invention, the method is applicable to an electronic device and includes the following steps. First, the current time is taken and the interval between the processor and the next waking time point is calculated. Again, release this processor before this interval is reached. In addition, when it is judged that the interval time is reached or the first notification signal of the processor is received, the first latest time is obtained to update the current time, and the current time is recorded as the reference time, and then the plurality of registered executions in the thread are executed. Check whether the current time meets the preset time condition of multiple registered threads included in the memory. In addition, when it is determined that the current time satisfies the preset time condition of a first registered thread in the registered threads, the first registered thread is awake and executed.

This case proposes an electronic device. According to an embodiment of the present disclosure, the electronic device includes a memory and a processor. The memory includes an operating system and a plurality of threads, wherein the operating system includes a thread manager. this The processor is used to execute this operating system, the thread manager and these threads. Furthermore, the thread manager takes the current time and calculates the interval between the processor and the next waking time. This thread manager releases the processor before the interval is reached. When the thread manager determines that the interval time has elapsed or receives the first notification signal from the processor, the thread manager obtains the first latest time to update the current time, records the current time as the reference time, and targets the plurality of memories. The registered thread is checked to check whether the current time meets the preset time conditions of these registered threads. The thread manager wakes up the first registered thread when it is determined that the current time satisfies the preset time condition of a first registered thread in the registered threads.

Based on the above, the embodiment of the present invention proposes a thread management method and an electronic device thereof. Registering a plurality of threads via the thread manager, planning a execution flow of the processor according to a corresponding preset time condition of the execution interval of each thread, and awakening the thread corresponding to the preset time condition to be awakened to increase The time the processor is released, which in turn saves power consumption.

In order to make the above features and advantages of the present invention more comprehensible, the following detailed description of the embodiments and the accompanying drawings.

FIG. 2 is a functional block diagram of an electronic device according to an embodiment of the present invention. Referring to FIG. 2 , the electronic device 20 includes a memory 210 and a processor 220 . The memory 210 is connected to the processor 220 for storing the operating system 211, the driver unit 213, the application unit 214, and the program list. The plurality of threads corresponding to the code of the unit 215 and the program unit 216 are processed and executed by the processor 220. In addition, the operating system 211 includes a thread manager 212 that can be executed by the processor 220 to manage the execution timing of the plurality of threads in the electronic device 20. Processor 220 is, for example, a central processing unit or a processor circuit having a plurality of combinations of processor cores.

More clearly, the thread manager 212 can register a plurality of threads in the electronic device 20, but not all threads in the electronic device 20 register with the thread manager 212. The thread manager 212 can include a database (not shown) for recording the preset time conditions corresponding to the registered thread. The thread manager 212 obtains the current tick, which can be the system time provided by the processor. The thread manager 212 utilizes the current time to calculate the interval between the processor 220 and the next waking time point. Again, the thread manager 212 causes the processor 220 to be in a released state before the interval time is reached. That is, the thread manager 212 releases the processor 220 before the interval time and does not execute the registered thread. When the processor 220 does not process the registered thread of the thread manager 212, the state is released, the processor 220 can go to sleep, or the processor 220 executes the unregistered thread.

When the thread manager 212 determines that the first notification signal (or the first notification information) from the processor 220 is received before the interval has elapsed, the thread manager 212 obtains a first latest time to update. The current time, and recording the current time as a reference time. The reference time is the last (or last) execution of the processor 220. The end time of any thread in the registered thread. Then, the thread manager 212 can separately check whether the updated current time meets a plurality of preset time conditions of the registered threads for the plurality of registered threads in the plurality of threads in the memory 210. In addition, when the thread manager 212 checks that the current time meets the preset time condition of a registered thread in the registered threads, the thread manager 212 wakes up the registered thread and processes it. The device 220 executes the first registered thread. For example, the preset time condition of the registered thread includes: an ideal interval of the registered thread, a maximum interval, and a minimum interval.

FIG. 3 is a flow chart showing a method of thread management according to an embodiment of the present invention. Referring to FIG. 2 and FIG. 3 simultaneously, the thread management method 300 starts at step S301. In step S301, the thread manager 212 obtains the current time (for example, using the preset function to obtain the system time), and calculates an interval between the processor 220 from now to the next waking time point. The specific thread manager 212 calculates the interval method, which will be described in detail with reference to FIG.

At step S302, during this interval, the thread manager 212 places the processor 220 in a released state to enable it to perform other tasks or to be in a power saving state (eg, to enter a sleep mode). Next, in step S303, the thread manager 212 determines whether a first notification signal from the processor 220 is received before the interval time is reached. The first notification signal is, for example, that the processor 220 is awakened by other unregistered threads. Generated notification A signal is used to notify the thread manager 212 that the processor 220 is not in sleep mode. In step S303, if the thread manager 212 has received the first notification signal, step S305 is performed after step S303; if the thread manager 212 does not receive the first notification signal, it is executed after step S303. Step S304.

In step S304, the thread manager 212 further determines whether the current time has reached the next wake-up time point of the above-mentioned calculated processor 220, that is, whether the interval time calculated in step S301 has been reached. In step S304, if the thread manager 212 determines that the interval has been reached, step S305 is performed after step S304; if the thread manager 212 determines that the next wake-up time point has not been reached, the processor 220 continues to be released. The state does not execute any of the registered threads, and after step S304, returns to step S303 for reconfirmation.

As can be seen from the above, when it is determined that the calculated next wake-up time point is reached or the first notification signal from the processor 220 is received, step S305 is performed. In this step S305, the processor 220 has started executing one or more threads in the memory 210, which may be registered or unregistered threads. Therefore, in order to update the execution time record of the processor 220, in step S305, the thread manager 212 obtains a first latest time to update the current time, and records the current time as a reference time, the reference time. Represents the last (or last) execution of this processor 220 to complete the end time of any of the registered threads (or last round tickcount).

Step S306 is continued after step S305. At step S306 The thread manager 212 determines whether there is a registered thread that has not been checked. In step S306, when the thread manager 212 determines that there is still a registered thread that has not been checked yet, step S307 is performed after step S306; when the thread manager 212 determines that there is no registered thread that has not been checked yet, Returning to step S301 after step S306, the interval time is recalculated and the processor 220 is returned to the released state again. .

In step S307, the thread manager 212 selects a registered thread from one or more registered threads that have not been checked to determine whether the current time satisfies the preset time condition of the registered thread. In step S307, when the thread manager 212 determines that the current time does not meet the preset time condition of the currently checked registered thread, then returns to step S305 after step S307 to update the current time and reset the reference time; The thread manager 212 determines that the current time has met the preset time condition of the currently registered registered thread, and then performs step S308 after step S307. In step S308, the registered thread is awakened by the thread manager 212, and the processor 220 executes the relevant steps of the registered thread (or the thread to perform the corresponding work).

From another point of view, the steps S305, S306, and S307 are generally regarded as respectively checking whether the current time satisfies the preset time condition of each of the plurality of registered threads recorded by the thread manager 212.

If the processor 220 is awakened to operate because the thread manager 212 has reached the interval, the detailed determination will be described using FIG. If the thread manager 212 receives the first notification signal from the processor 220 before the interval interval, the detailed judgment manner will be as shown in FIG. description.

4A-4C are flowcharts illustrating another thread management method according to an embodiment of the present disclosure. Referring to FIG. 2 and FIG. 4A to FIG. 4C simultaneously, in this embodiment, steps S301 to S308 of FIG. 4B are similar to steps S301 to S308 of FIG. 3, and thus the related detailed technical content may refer to FIG. These technical contents are not re-described.

Referring to FIG. 2 and FIG. 4A, in step S401, a thread manager 212 is created in the operating system 211. The code of the thread manager 212 is stored in the memory 210, and the processor 220 processes the thread. When the code of the related program of the manager 212 is coded, the thread manager 212 can schedule one or more threads that have been registered, and plan the usage time of the processor 220, thereby making the execution time of the registered thread as much as possible. The power saving mode period of the processor 220 is consistent. This can reduce the frequency at which the processor 220 is woken up by the power saving mode (or the sleep mode), or reduce the busy time of the processor 220.

The step of registering the registered threads may include using a table (described in Table 1 below), a data structure (not shown) in the memory 210, or a database in the memory 210 (not shown) ) Record the ideal interval (Interval(ideal)), the maximum allowable interval (Interval(max)), and the minimum allowable interval (Interval(min)) of these registered threads, in order to make the thread flexible. The difference between the maximum allowable interval and the minimum allowable interval recorded by the thread manager 212 is greater than or equal to the sleep mode unit time of the processor (or CPU_IDLE_REQUIRE_TIME).

In this embodiment, there are still threads in the memory that are not registered with the thread manager 212. These threads are not controlled or managed by the thread manager 212, so when one of these unregistered threads operates At this time, the processor 220 may be caused to issue a first notification signal to notify the thread manager 212 that the processor 220 is already in a busy state or is woken up by the sleep mode.

In step S403, the thread manager 212 determines whether there is a driver or application that needs to periodically read/write data or periodically perform corresponding work, or whether a new thread is generated and directed to the thread manager 212. Sign up. In step S403, if the thread manager 212 determines that the above event has not occurred (corresponding to the condition that the different thread is to be executed), the step S403 is repeatedly performed for reconfirmation; if the thread manager 212 determines that the event occurs Then, step S301 is performed after step S403, and the technical content of step S301 to step S308 in FIG. 4B is followed by FIG. 3. That is, since there is a thread in step S403 that needs to perform the corresponding work, or whether a new thread is generated and registered with the thread manager 212, the thread manager 212 needs to reacquire the current time after step S403. And calculating the interval between the current time point of the processor 220 and the next waking time point.

Referring to FIG. 2 and FIG. 4C, step S405 is performed after step S308. In step S405, the awakened registered thread determines whether or not the work is completed. In step S405, when the registered thread has completed the work, the step S407 is continued after the step S405; when the registered thread has not completed the work, the step S405 is repeatedly executed after the step S405 to reconfirm.

It should be noted that, in step S308, after a registered thread is invoked to perform the related action, the process proceeds to step S405 to repeatedly determine whether the thread completes the related action, or proceeds to step S405 after completing all related actions. If the related action has been completed, the registered thread will send a second notification signal to the thread manager 212 in step S407, and then proceed to step S301 to obtain a second latest time to update the current time and utilize The current time recalculates another interval between the processor 220 and the next waking time point.

An ideal interval time, a maximum allowable interval time, and a minimum allowable interval time of the registered threads that can be included in the preset time condition will be described in detail below with reference to Table 1 as an example.

Table 1 details the preset time conditions of Thread 1 to Thread 4. Referring to Table 1, the thread 1 to thread 4 of Table 1 are, for example, preset time conditions of the driver unit 213, the application unit 214, the program unit 215, and the program unit 216. These preset time conditions can be recorded when the thread manager 212 registers these threads and are then used to calculate the appropriate interval. Every other time.

The interval (ideal), interval (maximum) and interval (minimum) of thread 1 to thread 4 are all multiples of the sleep mode unit time of the processor. Only the sleep mode units are listed in Table 1. The multiple of the time (or multiplier) and the unit is milliseconds (ms). The last time tick is also rounded to a multiple of the sleep mode unit time. For example, the last execution time of thread 1 is 150 ms, and its ideal interval time is 5 times sleep mode unit time, the maximum allowable interval time is 8 times sleep mode unit time, and the minimum allowable interval time is 0 times sleep mode unit time, this is the minimum interval after executing thread 1 for processor 220 to immediately process thread 1 again.

The ideal interval time of the other threads 2 to Thread 4, the maximum allowable interval time, and the minimum allowable interval time can be known from Table 1. For example, assuming that the sleep mode unit time of the processor is 50 ms, the thread manager 212 can estimate from Table 1 that the ideal interval time for the next wake-up thread 1 is 150+50*5=400 ms; The maximum allowable interval for wake-up thread 1 is 150+50*8=550 ms; the minimum allowable interval for wake-up thread 1 from Table 1 is 150+50*0=150 ms. According to the above principle, the interval (ideal), interval time (maximum), and interval time (minimum) of the thread 2 to the thread 4 can be derived.

It is worth mentioning here that the last execution time of each thread in Table 1 is continuously updated by the thread manager 212 obtaining the latest current time.

FIG. 5 is a flow chart of a method for calculating a suitable interval time according to an embodiment of the present invention. Referring to FIG. 5, the calculation of the suitable interval time method details the technical content of step S301, and starts at step S501. In step S501, the thread manager 212 may obtain a second latest time to update the current time, and subtract the current time from the last time the processor 220 executed the end time of any registered thread (or the reference time). Taking a time difference as the first time difference.

Next, in step S502, the thread manager 212 performs a trimming operation on the first time difference value to adjust the first time difference value to a multiple of the sleep mode unit time to become the sleep mode unit time and the first The second time difference of the product of a multiplier. Here, the first multiplier is a magnification. For example, if the sleep mode unit time is 50, the trimming operation is in units of 25, and if the first time difference is greater than or equal to 25, the first multiplier is taken as 1, so the second time difference is 50. If the sleep mode unit time is 50, the trimming operation is in units of 25, and if the first time difference is less than 25, the first multiplier is 0 and the second time difference is 0. As such, the time difference is adjusted to a multiple of the sleep mode unit time so that the processor 220 has sufficient time to enter the sleep mode.

In step S503, the thread manager 212 adds the second time difference of the multiple of the sleep mode unit time to the reference time to obtain a third time difference, wherein the third time difference is just executed or just The latest execution time or registration time of one of the registered threads is added (refer to the timing of step S301 with reference to FIGS. 4A and 4C, that is, the registered thread registration time or the execution end time). Therefore, in step S504, execution is performed The manager 212 can also use the third time difference to update the latest registration time of one or more threads newly added from the reference time to the second latest time, or update from the reference time to the second latest The last execution time of one of the registered threads that has just been executed. As a result, the last registration time of some registered threads, or the last execution time, will be adjusted to the same time point.

In step S505, the thread manager 212 adds the updated registration time of the newly added thread (or the newly registered thread) or the latest execution time of the registered thread just executed, respectively, to the corresponding maximum tolerance. Interval and get multiple intervals.

In step S506, the thread manager 212 selects a minimum interval time value from among the interval time values as the next earliest wake-up time point of the processor, and updates the interval time accordingly. The interval is an interval between the current time and the next waking time point that the processor 220 wakes up next time. The thread manager 212 places the processor 220 in the released state during the interval. The detailed steps for calculating the interval are described below.

In step S507, the thread manager 212 subtracts the current time from the earliest wake-up time point to obtain an interval time as a fourth time difference, the fourth time difference being the current time and the processor 220 next time. A time difference between the time points to wake up. Next, in step S508, the thread manager 212 performs the fourth time difference trimming operation to become a multiple of the sleep mode unit time to obtain a fifth time difference value, wherein the fifth time difference value is the sleep mode unit time multiplication. The product value in the second multiplier.

In step S509, the thread manager 212 determines whether the second multiplier is zero. In step S509, when the thread manager 212 determines that the second multiplier is zero, step S510 is performed after step S509; in step S509, when the thread manager 212 determines that the second multiplier is not zero, Then, step S511 is performed after step S509.

In step S510, the thread manager 212 takes the fourth time difference value that has not undergone the trimming operation as the interval time. In step S511, the thread manager 212 takes the fifth time difference value of the trimming operation as the interval time to obtain the optimal time utilization of the processor 220.

FIG. 6 is a flow chart of a method for checking whether a current time meets a thread preset time condition according to an embodiment of the present invention. Referring to FIG. 2 and FIG. 6, at the same time, in this embodiment, when the interval time is reached, the processor 220 ends the released state. In step S601, the thread manager 212 obtains a plurality of time differences for each of the registered threads, respectively, by obtaining the current execution time minus the latest execution time of all the registered threads.

Next, in step S602, the thread manager 212 determines, for each of the registered threads, whether the calculated time differences are greater than or equal to a preset ideal interval time corresponding to the registered threads. That is, in step S602, the thread manager 212 determines whether the corresponding time difference value of the registered thread is greater than or equal to its corresponding preset ideal interval time. In step S602, if the thread manager determines that the corresponding time difference of a registered thread is greater than or equal to the corresponding preset ideal interval time, step S604 is performed after step S602. That is, execution The thread manager 212 determines that the current time meets the preset time condition of the registered thread. In step S604, the thread manager 212 wakes up the registered thread, and the processor 220 executes the corresponding action of the registered thread.

In step S602, if the thread manager 212 determines that the corresponding time difference of a registered thread is less than the corresponding preset ideal interval time, step S603 is performed after step S602. In step S603, the thread manager 212 determines that the current time does not meet the preset time condition of the registered thread.

FIG. 7 is a flow chart showing another method for checking whether the current time meets the thread preset time condition according to an embodiment of the present invention. Referring to FIG. 2 and FIG. 7, at the same time, in this embodiment, since the thread manager 212 receives the first notification signal from the processor 220, it is known that the processor 220 ends its release state. The first notification signal occurs, for example, when the user has an input action, the processor 220 is awakened by the thread that processes the input action, so the thread manager 212 needs to re-schedule the processor to wake up next time. The interval is such that the time the thread is expected to be executed is as close as possible to the interval at which the processor will wake up next time. Alternatively, synchronize the time that the thread is expected to be executed to the interval at which the processor will wake up next time. In step S701, the thread manager 212 obtains the current time, and subtracts the last execution time corresponding to the registered threads by the current time to obtain a plurality of time differences.

Next, in step S702, the thread manager 212 determines, for each of the registered threads, whether the time differences are greater than or equal to the corresponding minimum ideal interval of the registered threads. This is similar to In step S602, the thread manager 212 determines whether the corresponding time difference value of the registered thread is greater than or equal to its corresponding preset ideal interval time.

In step S702, if the thread manager 212 determines that the time difference value is less than the corresponding minimum ideal interval time of the registered threads, then step S703 is followed by step S703. That is, the thread manager 212 determines that the current time does not meet the preset time condition of the registered thread. In step S702, if the thread manager 212 determines that the time difference is greater than the corresponding minimum ideal interval of the registered threads, then step S704 is followed by step S704. In step S704, the thread manager 212 wakes up the registered thread, and the processor 220 executes the corresponding action of the registered thread.

In the embodiment of FIG. 7, when the thread manager 212 adjusts the time schedule of each registered thread to be consistent with the execution cycle of the processor 220 as soon as possible, the frequency hopping can be reduced, and thus In step S702, it is selected to determine whether the current time meets the preset time condition of the registered thread by using the corresponding minimum ideal interval time.

The present invention further provides a computer program product for storing a computer executable program for performing the steps of the above thread management method. The computer program product is basically composed of a plurality of code segments ( For example, creating an organization chart code segment, signing a form code segment, setting a code segment, and deploying a code segment, and the code segments are loaded into the communication device and executed, the thread management method can be completed. The steps are the same as those of the above electronic device.

In summary, various embodiments of the present invention propose a thread management method and an electronic device thereof. Registering multiple threads via the thread manager in the operating system, planning the execution flow of the processor according to the corresponding preset time condition of the execution interval of each thread, or flexibly waking up the processor in response to the non-registered thread , wake up the thread that meets the preset time condition in advance. In this way, the time when the processor is released can be increased, and the frequency of the processor entering the busy mode from the sleep mode can be reduced, thereby achieving the power consumption saving effect.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Anyone having ordinary knowledge in the technical field can protect the case without any deviation and refinement within the spirit and scope of the present case. The scope is subject to the definition of the scope of the patent application attached.

20‧‧‧Electronic devices

210‧‧‧ memory

211‧‧‧ operating system

212‧‧‧Thread Manager

213‧‧‧Driver unit

214‧‧‧Application unit

215, 216‧‧‧ program unit

220‧‧‧ processor

A1~A6, B1~B6‧‧‧ time points

S301~S308, S401~S407, S501~S511, S601~S604, S701~S704‧‧‧ steps

1A to 1C are schematic diagrams showing the execution timing of a thread and a processor.

FIG. 2 is a functional block diagram of an electronic device according to an embodiment of the present invention.

FIG. 3 is a flow chart showing a method of thread management according to an embodiment of the present invention.

4A-4C are flowcharts illustrating a thread management method according to an embodiment of the present disclosure.

FIG. 5 is a flow chart of a method for calculating a suitable interval according to an embodiment of the present invention.

FIG. 6 is a flow chart showing a method for checking that a current time meets a preset condition of a thread according to an embodiment of the present invention.

FIG. 7 is a flow chart showing another method for checking that the current time meets the thread preset condition according to an embodiment of the present invention.

S301~S308‧‧‧ steps of the thread management method of an embodiment of the present invention

Claims (22)

A thread management method is applicable to an electronic device, the electronic device includes a processor and a memory, the memory includes a plurality of threads, and the method includes: obtaining a current time, and calculating the processing An interval time from the next wake-up time point; before the interval time is reached, the processor is released; when it is determined that the interval time is reached or a first notification signal of the processor is received, a first The latest time is to update the current time, record the current time as a reference time, and check whether the current time meets multiple presets of the registered threads for the plurality of registered threads in the threads. a time condition; and waking up the first registered thread when the current time satisfies a preset time condition of a first registered thread in the registered threads. The thread management method of claim 1, wherein the thread management method further comprises: creating a thread manager before calculating the interval time of the processor from the next wake-up time point; And using the thread manager to register some of the threads in the threads for the registered threads. The thread management method of claim 2, wherein the step of determining that the interval time is reached or receiving the first notification signal comprises: determining whether the processor is received before the interval time is reached a first notification signal, wherein the processor informs the thread manager via the first notification signal that the processor is currently awake. The thread management method of claim 1, wherein the step of separately checking whether the current time meets the preset time conditions of the registered threads comprises: determining that the current time cannot meet the current check When the preset time condition of the first registered thread is selected, a second registered thread that has not been checked is selected, and a second latest time is obtained to update the current time. The thread management method of claim 1, wherein after the step of waking up the first registered thread, the thread management method further comprises: receiving a first registered thread When the signal is notified, a second latest time is obtained to update the current time, and another time interval between the processor and the next wake-up time point is recalculated, wherein the first registered thread is notified by the second notification signal. The thread manager has completed the execution of the first registered thread. The thread management method of claim 4, wherein the step of separately checking whether the current time meets the preset time conditions of the registered threads comprises: determining that the current time cannot satisfy the When the preset time conditions of the thread have been registered, the step of obtaining the current time and calculating the interval time of the processor from the next wake-up time point is re-executed. The thread management method according to claim 1, wherein the target is checked for a plurality of registered threads in the threads Whether the previous time satisfies the plurality of preset time conditions of the registered threads includes: when the interval time is reached, respectively obtaining the current time minus the corresponding last execution time of the registered threads to obtain a plurality of time difference values; and determining, for each of the registered threads, whether the time difference values are greater than or equal to a corresponding preset ideal interval of the registered threads; and when the number of the registered threads is the same When the corresponding time difference of a registered thread is greater than or equal to the corresponding preset ideal interval time, determining that the current time meets the corresponding preset time condition of the first registered thread; and when the registered threads are registered When the corresponding time difference of the first registered thread is less than the preset ideal interval time, it is determined that the current time does not satisfy the corresponding preset time condition. The thread management method of claim 1, wherein for the plurality of registered threads in the threads, checking whether the current time satisfies a plurality of preset time conditions of the registered threads The step of: receiving the first notification signal before the interval time is reached, respectively obtaining the current time minus the corresponding last execution time of the registered threads to obtain a plurality of time differences; The registered threads determine whether the time difference values are greater than corresponding preset minimum interval times of the registered threads; and corresponding to the first registered threads in the registered threads When the time difference is greater than or equal to the corresponding preset minimum interval time, determining that the current time meets the corresponding preset time condition of the first registered thread; and when the first registered thread in the registered threads When the corresponding time difference is less than or equal to the corresponding preset ideal interval time, it is determined that the current time does not satisfy the corresponding preset time condition. The thread management method of claim 2, wherein the step of registering the registered threads comprises: respectively recording an ideal interval time, a maximum allowable interval time, and a minimum tolerance of the registered threads. The interval time, wherein a difference between the maximum allowable interval time and the minimum allowable interval time is greater than or equal to a sleep mode unit time of the processor. The thread management method of claim 9, wherein the calculating the interval time comprises: obtaining a second latest time to update the current time, subtracting the current time from the current time to obtain a first time a time difference, wherein the reference time is an end time point at which the processor last executed any one of the registered threads; performing a trimming operation on the first time difference to obtain the sleep mode unit time multiplication The product of the previous first multiplier is a second time difference value; the second time difference value is added to the reference time to generate a third time difference value, and the third time difference value is respectively updated at the reference time The registration time of one or more new registration threads between the second most recent time or the last execution of a first registered thread in the registered threads Time; respectively, the last registration time of the new registration thread or the last execution time of the first registered thread plus the corresponding maximum allowable interval time to obtain multiple estimation intervals; and obtain the A minimum estimated time value in the estimated interval is the earliest wake-up time point that the processor wakes up next time. The thread management method according to claim 10, wherein the calculating the interval time further comprises: subtracting the current time from the earliest wake-up time point to obtain a fourth time difference, and the interval is The time is updated to the fourth time difference; the trimming operation is performed on the fourth time difference to obtain the product of the sleep mode unit time multiplied by a second multiplier as a fifth time difference; Whether the multiplier is 0; when the second multiplier is 0, updating the interval time is the fourth time difference; and when the second multiplier is greater than 0, updating the interval time to the fifth time difference . An electronic device comprising: a memory comprising an operating system and a plurality of threads, wherein the operating system includes a thread manager; and a processor for executing the operating system, the thread manager and the Threads, wherein: the thread manager obtains a current time and calculates an interval between the processor and the next waking time point; The thread manager releases the processor before the interval is reached; when the thread manager determines that the interval is reached or a first notification signal is received by the processor, the thread manager obtains a first The latest time to update the current time, record the current time as a reference time, and check whether the current time meets multiple presets of the registered threads for the plurality of registered threads in the threads. a time condition; and when the current time satisfies a preset time condition of a first registered thread in the registered threads, the thread manager wakes up the first registered thread. The electronic device of claim 12, wherein: the thread manager registers part of the threads in the threads as the registered threads; and the thread manager records the An ideal interval time, a maximum allowable interval time, and a minimum allowable interval time of the registration thread, wherein a difference between the maximum allowable interval time and the minimum allowable interval time is greater than or equal to a sleep mode unit time of the processor. The electronic device of claim 12, wherein the thread manager further determines whether the first notification signal of the processor is received before the interval is reached, wherein the processor is notified by the first notification signal The thread manager is currently waking up with the processor. The electronic device of claim 12, wherein the thread manager determines that the current time cannot satisfy the current check When a preset time condition of a registered thread is received, the thread manager selects a second registered thread that has not been checked, obtains a second latest time to update the current time, and recalculates the processor distance to wake up next time. Another interval of time to come. The electronic device of claim 12, wherein when the thread manager receives a second notification signal of the first registered thread, the thread manager obtains a second latest time to update At the current time, recalculating another interval of the processor from the next waking time point, wherein the first registered thread informs the thread manager via the notification signal that the first registered thread has completed its execute program. The electronic device of claim 15, wherein the thread manager obtains a first time when the thread manager determines that the current time cannot satisfy the preset time conditions of the registered threads. Three latest times to update the base time and recalculate another interval between the processor and the next wake-up time point. The electronic device of claim 12, wherein: when the interval is reached, the thread manager obtains the current time minus the corresponding last execution time of the registered threads to obtain more Time difference value; and the thread manager respectively determines, for the registered threads, whether the time differences are greater than corresponding preset ideal intervals of the registered threads; when the thread manager determines the The corresponding time difference of the first registered thread in the registered thread is greater than the corresponding preset ideal interval. During the time interval, the thread manager determines that the current time meets the corresponding preset time condition of the first registered thread; and when the thread manager determines the first registered thread in the registered threads When the corresponding time difference is less than or equal to its preset ideal interval time, the thread manager determines that the current time does not satisfy its corresponding preset time condition. The electronic device of claim 12, wherein: when the first notification signal is received before the interval is reached, the thread manager obtains the current time minus the correspondence of the registered threads The last execution time to obtain a plurality of time differences; and the thread manager respectively determines, for the registered threads, whether the time differences are greater than corresponding minimum ideal intervals of the registered threads; When the thread manager determines that the corresponding time difference of the first registered thread in the registered thread is greater than the corresponding preset ideal interval, the thread manager determines that the current time satisfies the first registered time. Corresponding preset time condition of the thread; and when the thread manager determines that the corresponding time difference of the first registered thread in the registered thread is less than or equal to the corresponding preset ideal interval time, the execution The administrator determines that the current time does not meet its corresponding preset time condition. The electronic device of claim 13, wherein the thread manager records an ideal interval time, a maximum allowable interval time, and a minimum allowable interval time of the registered threads, wherein the A difference between the maximum allowable interval time and the minimum allowable interval time is greater than or equal to a sleep mode unit time of the processor. The electronic device of claim 20, wherein: the thread manager obtains a second latest time to update the current time, subtracting the current time from the current time to obtain a first time difference, The reference time is an end time point of the last execution of the one of the registered threads by the processor; the thread manager performs a trimming operation on the first time difference to obtain the sleep mode unit time Multiplying the product of a first multiplier into a second time difference; the thread manager adds the second time difference to the reference time to generate a third time difference, using the third time difference Updating a registration time of one or more new registration threads between the reference time and the second latest time, or a last execution time of a first registered thread in the registered threads; the execution The administrator separately adds the last registration time of the new registration thread or the last execution time of the first registered thread to the corresponding maximum allowable interval time to obtain Estimating a time interval; and a point in time of the earliest wake manager thread acquired the next time the plurality of estimating a minimum time interval of the estimated value of the primary processor to wake up. The electronic device of claim 21, wherein: the thread manager subtracts the current time from the earliest wake-up time point to obtain a fourth time difference value, and updates the interval time to the first Four o'clock The difference is performed by the thread manager to perform the trimming operation on the fourth time difference to obtain a product of the sleep mode unit time multiplied by a second multiplier to a fifth time difference; the thread manager Determining whether the second multiplier is 0; when the second multiplier is 0, the thread manager updates the interval time to the fourth time difference value; and when the second multiplier is not 0, the The thread manager updates the interval to the fifth time difference.