CN114528080A - Multi-task scheduling method and device - Google Patents

Abstract

The embodiment of the invention discloses a multitask scheduling method and a device, wherein the method comprises the following steps: acquiring a task configuration file, wherein the task configuration file comprises configuration information of a plurality of target tasks, and the configuration information comprises a task scheduling period and a task scheduling sequence; the method comprises the steps of scheduling a plurality of target tasks and monitoring task states by using a daemon process based on a task configuration file, wherein a task scheduling clock of the daemon process is established according to a task scheduling period and a task scheduling sequence, each target task corresponds to one task scheduling clock, a shared memory corresponding to the daemon process is used for data exchange among the target tasks and storing first task state information, and the first task state information is generated by the target tasks in the process of executing scheduling. By applying the method and the device, unified scheduling of multiple tasks is realized.

Description

A kind of multitask scheduling method and device

technical field

The present invention relates to the field of computer technology, and in particular, to a multitask scheduling method and device.

Background technique

With the in-depth development of computer technology, the application of multitasking operating systems is becoming more and more common. Multitasking operating systems such as Windows, Linux, QNX (Quick UNIX), etc., often run multiple task processes at the same time in multitasking operating systems. In the prior art, multiple tasks are usually started and run independently. For example, when multiple tasks need to be started, whether by manual method or command script method, each task needs to be started one by one. The operation is time-consuming, especially. It is inconvenient for maintenance personnel when a large number of tasks need to be started at the same time. In addition, there is no coordination between tasks. If a task is abnormal, the execution status of other related tasks cannot be adjusted and controlled in time. How to uniformly allocate and manage multiple tasks in a multitasking operating system is a problem that needs to be solved in the application field of the multitasking operating system.

SUMMARY OF THE INVENTION

The invention provides a multi-task scheduling method and device to realize unified scheduling of multi-tasks. The specific technical solutions are as follows:

In a first aspect, an embodiment of the present invention provides a multitask scheduling method, including:

Obtain a task configuration file, wherein the task configuration file includes configuration information of multiple target tasks, and the configuration information includes a task scheduling period and a task scheduling sequence;

Based on the task configuration file, the daemon process is used to schedule and monitor the task status of multiple target tasks. The task scheduling clock of the daemon process is determined according to the task scheduling cycle and task scheduling sequence. Each target task corresponds to a task scheduling clock. The corresponding shared memory is used for data exchange between each target task and for storing first task state information, where the first task state information is task state information generated by each target task during the execution and scheduling process.

Optionally, based on the task configuration file, before using the daemon to schedule and monitor multiple target tasks, including:

For each target task, a task scheduling clock is determined according to the task scheduling period of each target task and the task scheduling order of the associated target task, wherein multiple task scheduling clocks corresponding to multiple target tasks have the same clock source;

Determine the shared memory according to the inter-task data information and the second task state information, and the capacity of the shared memory corresponds to the sum of the inter-task data information and the capacity of the second task state information;

For each target task, the task process is determined according to the configuration information corresponding to each target task, and the initialized task process is associated with the shared memory.

Optionally, for each target task, according to the task scheduling period of each target task and the task scheduling sequence of the associated target task, determine the task scheduling clock, including:

Determine the task scheduling reference clock according to the task scheduling periods of the multiple target tasks, wherein the task scheduling reference clock corresponds to the greatest common divisor of the task scheduling periods of the multiple target tasks;

Determine the scheduling cycle count value of the target task according to the task scheduling reference clock and the task scheduling cycle of each target task, wherein the scheduling cycle count value is N reference times corresponding to the task scheduling reference clock, and N is a positive integer;

Determine the count offset value of each target task according to the task scheduling sequence;

According to the task scheduling sequence, determine the task execution mode of each target task, wherein,

When the task execution mode of the target task is serial execution, the task scheduling clock of the target task corresponds to the sum of the period count value of the target task and the count offset value of the target task;

When the task execution mode of the target task is parallel execution, the task scheduling clocks of multiple target tasks executed in parallel are the same, and the task scheduling clocks of multiple target tasks executed in parallel correspond to the cycle count value of each target task in the parallel execution tasks Sum of count offsets from own target task.

Optionally, for each target task, the task scheduling clock is determined according to the task scheduling period of each target task and the task scheduling sequence associated with the target task, and further includes:

According to the task scheduling reference clock, the system timer is determined. The system timer is the smallest clock slice for all task scheduling. The timer is used to realize the synchronous scheduling of multiple target tasks based on the same clock source.

Optionally, the shared memory performs data reading and writing based on system process locks.

Optionally, based on the task configuration file, use the target daemon to schedule and monitor multiple target tasks, including:

Trigger the target task according to the task scheduling clock of the target task;

Based on the data offset address of each associated target task, obtain the first data associated with the target task stored in the shared memory, wherein the first data is the data associated with the target task to be sent to the target task;

Based on the first data, the target task is executed, and second data is determined, wherein the second data corresponds to the execution result of the target task, the second data includes third data and fourth data, and the third data is to be received associated with the target task data, the fourth data is the state information data determined after the target task executes the task;

According to the fourth data, the target task and its associated target task are monitored.

Optionally, monitor the target task and its associated target task according to the fourth data, including:

get the fourth data;

According to the fourth data, determine whether the execution result of the target task reports an error and/or whether the target task exits abnormally;

In response to an error report of the execution result of the target task and/or the abnormal exit of the target task, a corresponding control instruction is issued to the associated target task according to a preset error level.

Optionally, the task configuration information also includes at least one of the task name, task core binding information, task priority, second task status information, and inter-task data information, where the second task status information is each target task. Raw task status information.

In a second aspect, an embodiment of the present invention provides a multi-task scheduling apparatus, the apparatus includes:

an acquisition module, configured to acquire a task configuration file, wherein the task configuration file includes configuration information of multiple target tasks, and the configuration information includes a task scheduling period and a task scheduling sequence;

The scheduling module is used to schedule multiple target tasks and monitor task status by using the daemon process based on the task configuration file. The task scheduling clock of the daemon process is determined according to the task scheduling period and task scheduling sequence, and each target task corresponds to a task The scheduling clock and the shared memory corresponding to the daemon process are used for data exchange between target tasks and for storing first task status information, which is task status information generated by each target task during the scheduling process.

Optionally, the device further includes:

The first determination module is configured to, for each target task, determine the task scheduling clock according to the task scheduling period of each target task and the task scheduling sequence of the associated target task, wherein the multiple task scheduling clocks corresponding to the multiple target tasks have the same clock source.

The second determining module is configured to determine the shared memory according to the inter-task data information and the second task state information, and the capacity of the shared memory corresponds to the sum of the inter-task data information and the capacity of the second task state information.

The third determining module is configured to, for each target task, determine the task process according to the configuration information corresponding to each target task, and associate the initialized task process with the shared memory.

Optionally, the first determining module may further include:

The first determining unit is configured to determine the task scheduling reference clock according to the task scheduling cycles of the multiple target tasks, wherein the task scheduling reference clock corresponds to the greatest common divisor of the task scheduling cycles of the multiple target tasks.

The second determining unit is configured to determine the scheduling cycle count value of the target task according to the task scheduling reference clock and the task scheduling cycle of each target task, wherein the scheduling cycle count value is N reference times corresponding to the task scheduling reference clock, N is a positive integer.

The third determining unit is configured to determine the count offset value of each target task according to the task scheduling sequence.

The fourth determination unit is used to determine the task execution mode of each target task according to the task scheduling sequence, wherein, when the task execution mode of the target task is serial execution, the task scheduling clock of the target task corresponds to the target task's The sum of the cycle count value and the count offset value of the target task; when the task execution mode of the target task is parallel execution, the task scheduling clock of multiple target tasks executed in parallel is the same, and the tasks of multiple target tasks executed in parallel The scheduling clock corresponds to the sum of the cycle count value of each target task in the parallel execution tasks and the count offset value of its own target task.

Optionally, the first determining module may further include:

The second determining unit is used for determining a system timer according to the task scheduling reference clock. The system timer is the smallest clock slice for all task scheduling, and the timer is used to realize synchronous scheduling of multiple target tasks based on the same clock source.

In an optional embodiment, the shared memory performs data reading and writing based on system process locks.

In an optional embodiment, the task configuration information further includes at least one of a task name, task core binding information, task priority, second task status information, and inter-task data information, wherein the second task status information Raw task status information for each target task.

Optionally, the scheduling module can include:

The first triggering unit is configured to trigger the target task according to the task scheduling clock of the target task.

The first obtaining unit is configured to obtain the first data of the associated target task stored in the shared memory based on the data offset address of each associated target task, where the first data is data to be sent to the target task by the associated target task.

a first determining unit, configured to execute the target task based on the first data, and determine second data, wherein the second data corresponds to the execution result of the target task, the second data includes third data and fourth data, and the third data is the data to be received by the associated target task, and the fourth data is the state information data determined after the target task executes the task.

The first monitoring unit is configured to monitor the target task and its associated target task according to the fourth data.

Optionally, the first monitoring unit may include:

The first obtaining subunit is used to obtain fourth data.

The first judging subunit is used for judging whether the execution result of the target task reports an error and/or whether the target task exits abnormally according to the fourth data.

The first response subunit is configured to issue a corresponding control instruction to the associated target task according to a preset error level in response to an error report of the execution result of the target task and/or the abnormal exit of the target task.

As can be seen from the above, a multi-task scheduling method and device provided by the embodiments of the present invention obtain a task configuration file, wherein the task configuration file includes configuration information of multiple target tasks, and the configuration information includes a task scheduling period and a task scheduling sequence; Based on the task configuration file, the daemon process is used to schedule and monitor the task status of multiple target tasks. The task scheduling clock of the daemon process is determined according to the task scheduling cycle and task scheduling sequence. Each target task corresponds to a task scheduling clock. The corresponding shared memory is used for data exchange between each target task and for storing first task state information, where the first task state information is task state information generated by each target task during the execution and scheduling process.

By applying the embodiments of the present invention, unified scheduling and real-time monitoring of multiple tasks can be performed. Of course, it is not necessary for any product or method of the present invention to achieve all of the advantages described above at the same time.

The technical effects of the embodiments of the present invention include:

1. In the prior art, each task in a multitasking operating system is started separately and runs independently. The embodiment of the present invention utilizes a daemon process to perform unified scheduling of multitasking, which can improve the efficiency of multitasking startup and control multiple tasks in real time. The embodiment of the present invention adds the role of unified allocation and management of multitasking in the multitasking operating system.

2. In the embodiment of the present invention, multiple tasks are scheduled in a configurable manner by a daemon process, and the daemon process generates a task schedule corresponding to each target task based on the same clock source according to the configuration information of each target task in the obtained task configuration file. The clock triggers the corresponding target task through the task scheduling clock, and realizes the clock synchronization between the target tasks.

3. In complex engineering applications, multi-tasks need to work collaboratively. For example, in multi-core distributed system applications, data communication is required between tasks. In the embodiment of the present invention, the target task and the associated target task use shared memory for data exchange, The joint application of each target task in engineering application is realized.

4. Each time the target task in the embodiment of the present invention performs task scheduling, the generated task status information is written into the corresponding position of the shared memory. Monitor and control the corresponding associated target tasks in real time according to the task status information of the target tasks, so as to realize the real-time monitoring and management of each target task.

Description of drawings

In order to illustrate the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that are required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

1 is a schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention;

FIG. 2 is another schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention;

FIG. 3 is another schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention;

FIG. 4 is another schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a multitask scheduling apparatus provided by an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

It should be noted that the terms "comprising" and "having" and any modifications thereof in the embodiments of the present invention and the accompanying drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the steps or units listed, but optionally also includes steps or units not listed, or optionally also includes For other steps or units inherent to these processes, methods, products or devices.

The present invention provides a multitask scheduling method and device. The embodiments of the present invention will be described in detail below.

FIG. 1 is a schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention. The method may include the following steps:

S101: Acquire a task configuration file, wherein the task configuration file includes configuration information of multiple target tasks, and the configuration information includes a task scheduling period and a task scheduling sequence.

The task configuration file is a predefined configuration file, and the task configuration file describes the process configuration information of each task in each engineering application. Wherein, each engineering application may include multiple tasks, and each task may be associated with multiple other tasks.

The task scheduling period refers to the scheduling period of each target task, that is, the time interval for triggering the scheduling of a target task, and each target task corresponds to a task scheduling period. The task scheduling order refers to the order in which each target task is scheduled. The task scheduling order is divided into two categories, one is parallel scheduling, and the other is serial scheduling. The scheduling order of each task in parallel scheduling is the same, and the scheduling order of each task in serial scheduling is the same. There is a sequence.

S102: Based on the task configuration file, use a daemon process to schedule multiple target tasks and monitor task status, wherein the task scheduling clock of the daemon process is determined according to the task scheduling period and the task scheduling order, and each The target task corresponds to a task scheduling clock, and the shared memory corresponding to the daemon process is used for data exchange between each target task and for storing first task status information, where the first task status information is the execution schedule of each target task. During the process, the task status information is generated.

The embodiment of the present invention utilizes a daemon process to dynamically schedule each target task and monitor the real-time task status. The daemon process is automatically started when the multitasking operating system is started, and automatically exits when the operating system is shut down. After each start of the daemon process, the task configuration file can be read once, and multiple target tasks can be scheduled according to the read task configuration file. The daemon process in the embodiment of the present invention is created based on the task configuration file, so the task scheduling clock of the daemon process can be created according to the task scheduling period and task scheduling sequence of each target task in the task configuration file. The task scheduling clocks of tasks are different, and the task scheduling clocks of multiple tasks scheduled in parallel are the same. Corresponding to the daemon process, there is also a shared memory. The shared memory is used to store the data communication information and the first task state information generated after each target task executes the task, and use the shared memory to exchange data between the target tasks.

In the prior art, the multitasking operating system lacks a role of task configuration and management monitoring, and it is difficult to achieve coordination and synchronization among multiple tasks. By applying the embodiments of the present invention, unified scheduling of multitasking can be realized, and the daemon process can be implemented. After startup, suspend multiple tasks at the same time and wait for the trigger of the task scheduling clock. According to the task execution result generated after triggering the target task, it can monitor its own target task and control the associated target task in real time.

In an optional embodiment, the task configuration information further includes at least one of a task name, task core binding information, task priority, second task status information, and inter-task data information, wherein the first The second task state information is the original task state information of each target task.

The task name refers to the name of each target task in each engineering application. The task core binding information is information specifying that each target task is bound to be executed on a certain core of the processor. In this embodiment of the present invention, the target task may be specified to a specific core, or it may not be specified by default. Automatic allocation, that's all possible. The task priority is to specify the priority of each target task process. It can be specified within the operating system process priority range of 0-255, or it can be not specified by default and inherit the daemon process priority, which is not limited here. The second task status information refers to task status feedback information corresponding to each target task before applying the guardian in the embodiment of the present invention to schedule each target task, including but not limited to the offset address of the original status data. The inter-task data information represents the data information exchanged between each target task, including but not limited to the read and write data offset addresses, data lengths, and the like.

FIG. 2 is another schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention. On the basis of the above embodiment, before using the daemon process to schedule and monitor multiple target tasks based on the task configuration file, as shown in FIG. 2 , it may further include:

S201: For each target task, determine the task scheduling clock according to the task scheduling period of each target task and the task scheduling sequence associated with the target task, wherein a plurality of the tasks corresponding to the plurality of target tasks Scheduled clocks have the same clock source.

Each target task has a task scheduling clock corresponding to itself, and all task scheduling clocks are based on the same clock source, so the clock synchronization between each target task is realized. It should be noted that the generation of the task scheduling clock of the target task is not only based on the task scheduling period of the target task and the task scheduling order associated with the target task, but also according to the task scheduling order of the target task. The associated target task in the embodiment of the present invention refers to one or more tasks associated with the target task.

S202: Determine the shared memory according to the inter-task data information and the second task state information, where the capacity of the shared memory corresponds to the sum of the inter-task data information and the capacity of the second task state information. .

In the embodiment of the present invention, a shared memory shared by each target task needs to be created, so that each target task can exchange data according to the offset address in the shared memory, and at the same time, it is also used for the daemon process to monitor and manage the running status information of each target task. Therefore, the size of the shared memory can be determined according to the total capacity of the inter-task data information and the second task state information in the task configuration file.

S203: For each target task, determine a task process according to the configuration information corresponding to each target task, and associate the initialized task process with the shared memory.

Before scheduling each target task, the process of each target task needs to be created. The corresponding task process can be created according to the configuration information of each target task in the task configuration file. In an achievable manner, first, according to the task name in the configuration information, find the task executable file corresponding to the target task; then, according to the core binding information in the configuration information, call a system command to bind the target task to the processor Then, according to the task priority in the configuration information, set the priority of the task process. After the task is created, the internal initialization of the task is performed, and it is connected to the created shared memory, and the target task is suspended to wait for the clock trigger signal. After receiving the clock trigger signal sent by the daemon process, execute a scheduling, after the execution is completed, it is in a suspended waiting state again, waiting for the arrival of the next clock trigger signal of the daemon process, so as to realize the daemon process scheduling each target task control.

In an optional embodiment, the shared memory performs data reading and writing based on system process locks.

In the embodiment of the present invention, the application system process lock coordinates the data read and write operations of each target task, prevents the read and write conflict of the shared memory data, and thus ensures the consistency of the data.

The creation of the daemon process mainly includes three aspects: the creation of the task scheduling clock, the creation of the corresponding shared memory, and the creation of each target task process. During the initialization phase, the daemon process reads the generated task configuration file and stores the configuration information of each target task. Read into an array of memory structures and store them for subsequent use. According to the stored task configuration information, create a task scheduling clock, share memory, and create each target task. During the running process, when the daemon process is started, each target task suspends each target task through the corresponding task process and waits for triggering. After being triggered by the task scheduling clock, the execution result is stored in the shared memory, and each target task is executed through the shared memory. Data exchange between target tasks, and finally monitor and control each target task according to the task status information stored in the shared memory.

The daemon process created by the application of the embodiment of the present invention can perform unified scheduling of multiple tasks, monitor each target task in real time, manage and control each target task according to the state information of each target task, and realize clock synchronization between each target task, and more Conducive to real-time control of associated target tasks.

FIG. 3 is another schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention. On the basis of the above embodiment, specifically, for each target task, the task scheduling clock is generated according to the task scheduling period of each target task and the task scheduling sequence associated with the target task, as shown in FIG. 3, can also include:

S301: Determine a task scheduling reference clock according to the task scheduling periods of multiple target tasks, where the task scheduling reference clock corresponds to the greatest common divisor of the task scheduling periods of the multiple target tasks.

Before determining the task scheduling clock of each target task, first determine the task scheduling reference clock. The task scheduling reference clock can be the greatest common divisor of the task scheduling cycles of all target tasks, and the greatest common divisor is taken instead of any common divisor. Conducive to simplifying the timing procedure.

S302: Determine a scheduling period count value of the target task according to the task scheduling reference clock and the task scheduling period of each target task, where the scheduling period count value is N corresponding to the task scheduling reference clock Base time, N is a positive integer.

Because one task scheduling cycle of the target task is N reference times corresponding to the reference clock, and the task scheduling cycle count value is the value of a single task scheduling cycle, the scheduling cycle count value is N reference times.

S303: Determine a count offset value of each target task according to the task scheduling sequence.

Calculate the count offset value according to the different scheduling sequence requirements of each target task. The count offset value is used to distinguish the sequential position of each target task in a single task scheduling trigger cycle. The task scheduling clock of each target task is obtained by adding a count offset value corresponding to the scheduling sequence of the target task on the basis of the task scheduling period of the target task. For example, three serial tasks (a/b/c) are scheduled once every 10ms, but the count offset value of task a is 1ms, the count offset value of task b is 3ms, and the count offset value of task c is 7ms, Then the a/b/c task scheduling clocks can be 11ms/13ms/17ms, 21ms/23ms/27ms, 31ms/33ms/37ms...; for another example, three parallel tasks (d/e/f) are scheduled once every 10ms , d/e/f task count offset values are all 5ms, then the d/e/f task scheduling clocks can be 15ms/15ms/15ms, 25ms/25ms/25ms, 35ms/35ms/35ms... respectively.

S304: Determine the task execution mode of each target task according to the task scheduling sequence, wherein, when the task execution mode of the target task is serial execution, the task scheduling clock of the target task corresponds to the period of the target task The sum of the count value and the count offset value of the target task; when the task execution mode of the target task is parallel execution, the task scheduling clocks of the multiple target tasks executed in parallel are the same, and the multiple targets executed in parallel have the same task scheduling clock. The task scheduling clock of a task corresponds to the sum of the cycle count value of each target task in the parallel execution tasks and the count offset value of its own target task.

Task scheduling sequence is divided into two kinds of serial scheduling, parallel scheduling. If some target tasks have serial execution requirements, then a corresponding count offset value needs to be added to the task cycle count value of each target task. Therefore, the timing of sending the clock trigger signal is controlled according to the cycle count value + the count offset value, which controls the order of task execution and realizes the serial execution of multiple tasks; if some target tasks are executed in parallel, then these The cycle count value + the count offset value of the target tasks are the same, that is, the task trigger clocks of multiple tasks executed in parallel are the same. The clock trigger signal in the above is also defined in the process of creating the task scheduling clock. The configuration information in the task configuration file may also include a task sequence number, and the clock trigger signal of each target task is defined according to the task sequence number. According to the task scheduling clock of each target task, the corresponding clock trigger signal is sent to the target task, and the target task is awakened to execute the task once. The clock trigger signal for each target task is different.

In an optional embodiment, for each target task, generating the task scheduling clock according to the task scheduling period of each target task and the task scheduling sequence associated with the target task may further include:

According to the task scheduling reference clock, a system timer is generated, where the system timer is the smallest clock slice for all task scheduling, and the timer is used to implement synchronous scheduling of multiple target tasks based on the same clock source.

All target tasks scheduled by the daemon are based on the same clock source, that is, all target tasks rely on the same system timer for timing. When the period count value of a target task is reached, add the time of the corresponding count offset value, Trigger the clock trigger signal of the target task once and send it to the responding task. After each target task process is created and started, the target task is in a state of suspending and waiting for a trigger signal. When a target task receives a clock trigger signal sent by the daemon process, a task scheduling is performed.

FIG. 4 is another schematic flowchart of a multitask scheduling method provided by an embodiment of the present invention. On the basis of the above embodiment, specifically, based on the task configuration file, the target daemon process is used to schedule and monitor multiple target tasks, as shown in FIG. 4 , may also include:

S401: Trigger the target task according to the task scheduling clock of the target task.

The daemon periodically schedules each target task. In each scheduling, first the daemon sends a clock trigger signal to the target task according to the task scheduling clock of the target task, so as to trigger the target task.

S402: Based on the data offset address of each associated target task, obtain first data of the associated target task stored in the shared memory, where the first data is data to be sent to the target task by the associated target task.

When the target task receives the trigger signal, it first reads the first data stored in the shared memory by its associated task from the shared memory. After each target task executes the task, the data that needs to be read by the associated target task is stored in the shared memory, and the part of the data that needs to be read by the associated target task is the first data, that is, the communication data between the associated target tasks. The first data is stored in a position corresponding to each associated target task, and the first data can be acquired through a data offset address corresponding to each associated target task.

S403: Execute the target task based on the first data, and determine second data, wherein the second data corresponds to the execution result of the target task, the second data includes third data and fourth data, and the The third data is data to be received by the associated target task, and the fourth data is state information data determined after the target task executes the task.

After the communication data of each associated target task is acquired, the target task is executed, and the data of the execution result, that is, the second data, is generated. The third data is the communication data related to the target task that needs to be read after the associated target task is triggered next time and before the task is executed. In the embodiment of the present invention, the shared memory is used to perform data exchange among the target tasks, thereby improving the data exchange efficiency among the target tasks.

S404: Monitor the target task and its associated target task according to the fourth data.

The daemon process can monitor the real-time status of the target task by acquiring the fourth data stored in the shared memory, and determine control instructions of multiple tasks associated with the target task according to the acquired real-time status information of the target task.

In an optional embodiment, the monitoring of the target task and its associated target task according to the fourth data may further include:

S501: Acquire the fourth data.

Each time a task scheduling is performed, each target task writes its own execution state to the corresponding location in the shared memory. The daemon can obtain real-time execution status information of each target task from the shared memory, so as to monitor each target task.

S502: According to the fourth data, determine whether the execution result of the target task reports an error and/or whether the target task exits abnormally.

The daemon process can know whether the execution result reports an error, the task execution is interrupted, and the abnormal exit occurs from the obtained task status information.

S503: In response to an error report of the execution result of the target task and/or the abnormal exit of the target task, issue a corresponding control instruction to the associated target task according to a preset error level.

The daemon can control the execution of other target tasks associated with the target task according to a predefined error level. For example, if task A and task B need to be executed serially, the serial execution method is to execute task A first, and then execute task B. If after scheduling task A, the execution result of task A shows that task A has exited abnormally. Then triggering task B is meaningless, so the daemon can issue an instruction to stop triggering task B. The daemon process in the embodiment of the present invention realizes collaborative management among multiple tasks.

Corresponding to the above method embodiments, the embodiments of the present invention provide a multi-task scheduling apparatus.

FIG. 5 is a schematic structural diagram of a multitask scheduling apparatus provided by an embodiment of the present invention. As shown in FIG. 5 , the apparatus may include:

an obtaining module 501, configured to obtain a task configuration file, wherein the task configuration file includes configuration information of multiple target tasks, and the configuration information includes a task scheduling period and a task scheduling sequence;

The scheduling module 502 is configured to use a daemon process to schedule multiple target tasks and monitor task status based on the task configuration file, wherein the task scheduling clock of the daemon process is based on the task scheduling period and the task scheduling sequence. It is determined that each target task corresponds to a task scheduling clock, and the shared memory corresponding to the daemon process is used for data exchange between each target task and for storing first task status information, where the first task status information is each target task. Task status information generated during the scheduling process of the task.

In an optional embodiment, the task configuration information further includes at least one of a task name, task core binding information, task priority, second task status information, and inter-task data information, wherein the first The second task state information is the original task state information of each target task.

In an optional embodiment, the apparatus further includes:

The first determining module is configured to, for each target task, determine the task scheduling clock according to the task scheduling period of each target task and the task scheduling sequence associated with the target task, wherein the corresponding A plurality of the task scheduling clocks have the same clock source.

A second determining module, configured to determine the shared memory according to the inter-task data information and the second task status information, where the capacity of the shared memory corresponds to the inter-task data information and the second task status information the sum of the capacity.

The third determining module is configured to, for each target task, determine a task process according to the configuration information corresponding to each target task, and associate the initialized task process with the shared memory.

In an optional embodiment, the first determining module may further include:

The first determining unit is configured to determine a task scheduling reference clock according to the task scheduling periods of multiple target tasks, wherein the task scheduling reference clock corresponds to the greatest common divisor of the task scheduling periods of the multiple target tasks.

a second determining unit, configured to determine the scheduling cycle count value of the target task according to the task scheduling reference clock and the task scheduling cycle of each target task, wherein the scheduling cycle count value is the task scheduling reference N reference times corresponding to the clock, where N is a positive integer.

The third determining unit is configured to determine the count offset value of each target task according to the task scheduling sequence.

The fourth determination unit is configured to determine the task execution mode of each target task according to the task scheduling sequence, wherein, when the task execution mode of the target task is serial execution, the task scheduling clock of the target task corresponds to the target task The sum of the cycle count value of the task and the count offset value of the target task; when the task execution mode of the target task is parallel execution, the task scheduling clocks of multiple target tasks executed in parallel are the same, and the parallel The task scheduling clock of the executed multiple target tasks corresponds to the sum of the cycle count value of each target task in the parallel execution tasks and the count offset value of its own target task.

In an optional embodiment, the first determining module may further include:

The second determination unit is configured to determine a system timer according to the task scheduling reference clock, where the system timer is the smallest clock slice scheduled by all tasks, and the timer is used to realize synchronization of multiple target tasks based on the same clock source schedule.

In an optional embodiment, the shared memory performs data reading and writing based on system process locks.

In an optional embodiment, the scheduling module may include:

The first triggering unit is configured to trigger the target task according to the task scheduling clock of the target task.

a first obtaining unit, configured to obtain the first data of the associated target task stored in the shared memory based on the data offset address of each associated target task, wherein the first data is the associated target task to be sent to the target task data.

The first determining unit is configured to execute the target task based on the first data, and determine second data, wherein the second data corresponds to the execution result of the target task, and the second data includes the third data and the first data. Four data, the third data is data to be received associated with the target task, and the fourth data is state information data determined after the target task executes the task.

The first monitoring unit is configured to monitor the target task and its associated target task according to the fourth data.

In an optional embodiment, the first monitoring unit may include:

The first obtaining subunit is used to obtain the fourth data.

The first judging subunit is used for judging whether the execution result of the target task reports an error and/or whether the target task exits abnormally according to the fourth data.

The first response subunit is configured to issue a corresponding control instruction to the associated target task according to a preset error level in response to an error report of the execution result of the target task and/or the abnormal exit of the target task.

Corresponding to the foregoing method embodiments, embodiments of the present invention provide a computer-readable storage medium for storing a computer program, wherein, when the computer program is executed by a processor, the multi-task scheduling method in the foregoing method embodiments is implemented .

The foregoing system and device embodiments correspond to the system embodiments, and have the same technical effects as the method embodiments. For specific descriptions, refer to the method embodiments. The apparatus embodiment is obtained based on the method embodiment, and the specific description can refer to the method embodiment section, which will not be repeated here. Those of ordinary skill in the art can understand that the accompanying drawing is only a schematic diagram of an embodiment, and the modules or processes in the accompanying drawing are not necessarily necessary to implement the present invention.

Those skilled in the art may understand that: the modules in the apparatus in the embodiment may be distributed in the apparatus in the embodiment according to the description of the embodiment, and may also be located in one or more apparatuses different from this embodiment with corresponding changes. The modules in the foregoing embodiments may be combined into one module, or may be further split into multiple sub-modules.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions in the embodiments of the present invention.

Claims (10)

1. A method for multitask scheduling, the method comprising:

acquiring a task configuration file, wherein the task configuration file comprises configuration information of a plurality of target tasks, and the configuration information comprises a task scheduling period and a task scheduling sequence;

and scheduling and monitoring the task state of a plurality of target tasks by using a daemon process based on the task configuration file, wherein a task scheduling clock of the daemon process is determined according to the task scheduling period and the task scheduling sequence, each target task corresponds to one task scheduling clock, a shared memory corresponding to the daemon process is used for data exchange among the target tasks and storing first task state information, and the first task state information is task state information generated by each target task in the process of executing scheduling.

2. The method of claim 1, wherein the task configuration information further comprises at least one of a task name, task core binding information, task priority, second task state information, and inter-task data information, wherein the second task state information is original task state information of each target task.

3. The method of claim 2, wherein prior to scheduling and monitoring the plurality of target tasks with the daemon process based on the task profile, comprising:

for each target task, determining the task scheduling clock according to the task scheduling period of each target task and the task scheduling sequence of the associated target task, wherein the task scheduling clocks corresponding to the target tasks have the same clock source;

and determining the shared memory according to the inter-task data information and the second task state information, wherein the capacity of the shared memory corresponds to the sum of the capacities of the inter-task data information and the second task state information.

And aiming at each target task, determining a task process according to the configuration information corresponding to each target task, and associating the initialized task process to the shared memory.

4. The method of claim 3, wherein the determining the task scheduling clock for each target task according to the task scheduling period for each target task and the task scheduling order for the associated target task comprises:

determining a task scheduling reference clock according to the task scheduling periods of the target tasks, wherein the task scheduling reference clock corresponds to the greatest common divisor of the task scheduling periods of the target tasks;

determining a scheduling period count value of each target task according to the task scheduling reference clock and the task scheduling period of the target task, wherein the scheduling period count value is N reference times corresponding to the task scheduling reference clock, and N is a positive integer;

determining a counting deviation value of each target task according to the task scheduling sequence;

determining a task execution mode of each target task according to the task scheduling sequence, wherein,

when the task execution mode of the target task is serial execution, the task scheduling clock of the target task corresponds to the sum of the period count value of the target task and the count offset value of the target task;

when the task execution mode of the target task is parallel execution, the task scheduling clocks of the multiple target tasks which are executed in parallel are the same, and the task scheduling clocks of the multiple target tasks which are executed in parallel correspond to the sum of the cycle count value of each target task in the tasks which are executed in parallel and the count offset value of the target task.

5. The method of claim 3, wherein the determining, for each target task, the task scheduling clock according to the task scheduling period for each target task and the task scheduling order for the associated target task, further comprises:

and determining a system timer according to the task scheduling reference clock, wherein the system timer is the minimum clock slice scheduled by all tasks, and the timer is used for realizing synchronous scheduling of a plurality of target tasks based on the same clock source.

6. The method of claim 3, wherein the shared memory reads and writes data based on a system process lock.

7. The method of claim 1, wherein the scheduling and monitoring of the plurality of target tasks with the target daemon based on the task profile comprises:

triggering the target task according to the task scheduling clock of the target task;

acquiring first data of the associated target task stored in the shared memory based on the data offset address of each associated target task, wherein the first data is data to be sent to the target task by the associated target task;

executing a target task based on the first data, and determining second data, wherein the second data corresponds to a target task execution result, the second data comprises third data and fourth data, the third data is data to be received by the associated target task, and the fourth data is state information data determined after the target task executes the task;

and monitoring the target task and the related target task according to the fourth data.

8. The method of claim 7, wherein monitoring the target task and its associated target task according to the fourth data comprises:

acquiring the fourth data;

judging whether the execution result of the target task is error-reported and/or whether the target task exits abnormally according to the fourth data;

and responding to the error report of the target task execution result and/or the abnormal exit of the target task, and issuing a corresponding control instruction to the associated target task according to a preset error level.

9. A multitask scheduling device, characterized in that said device comprises:

the system comprises an acquisition module, a task configuration module and a task scheduling module, wherein the acquisition module is used for acquiring a task configuration file, the task configuration file comprises configuration information of a plurality of target tasks, and the configuration information comprises a task scheduling period and a task scheduling sequence;

and the scheduling module is used for scheduling and monitoring the task states of a plurality of target tasks by using a daemon process based on the task configuration file, wherein a task scheduling clock of the daemon process is determined according to the task scheduling period and the task scheduling sequence, each target task corresponds to one task scheduling clock, a shared memory corresponding to the daemon process is used for data exchange among the target tasks and is used for storing first task state information, and the first task state information is task state information generated by each target task in the process of executing scheduling.

10. The apparatus of claim 9, wherein the apparatus further comprises:

the first determining module is used for determining the task scheduling clock according to the task scheduling period of each target task and the task scheduling sequence of the associated target task aiming at each target task, wherein a plurality of task scheduling clocks corresponding to a plurality of target tasks have the same clock source;

a first determining module, configured to determine the shared memory according to the inter-task data information and the second task state information, where a capacity of the shared memory corresponds to a sum of capacities of the inter-task data information and the second task state information;

and the first determining module is used for determining a task process according to the configuration information corresponding to each target task and associating the initialized task process to the shared memory aiming at each target task.

Images