JPH0934514A - Multitask control system for programmable controller and multitask type programmable controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten the time necessary for task switching and to improve the processing speed of a programmable controller. SOLUTION: When the tasks X1 to Xn receive the start instructions (a) from a system program SP (S3), these tasks start execution of their processing. Then every task carries on its processing until the information to be saved does not exist when a stop instruction (b) is received after a prescribed time T (S4). When no information to be saved exists, the task execution is stopped and a task request (c) is outputted to a system program SP (S5). The program SP instructs the start of the next task when the request (c) is received from one of tasks X1 to Xn that received the instruction (b) (S6, S2).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multitasking control system and a multitasking programmable controller for a programmable controller that processes a sequence program in a multitasking system.

[0002]

2. Description of the Related Art Conventionally, a programmable controller (hereinafter referred to as a sequencer) that processes a sequence program in a multitasking system has been proposed. That is, the sequence program is divided into a plurality of tasks and described (or, a plurality of sequence programs may be simultaneously executed with the sequence program as one task), and a part of the system program (operating system) By instructing execution and stop of each task by the control program, the processing is performed while switching the execution tasks, and the tasks are apparently processed in parallel. Since each task is time slice controlled, the execution task is forced at a relatively short time interval (usually a fixed time) so that one task does not occupy the central processing unit (CPU) for a long time. Can be switched. Such processing instructions are issued in the dedicated language of the sequencer (currently, the dedicated BASI
C is used).

FIG. 3 shows a processing procedure in the above-mentioned central processing unit (that is, a system program) of the multitask system.
(A). Now, assume that the number of tasks is m and the task number is represented by n. Further, it is assumed that the processing order of each task is fixed and the execution time of each task is a fixed time. When the processing of the sequence program is started, the task number n is first initialized to 0 (S
1). Next, the task number is incremented and the task number is set to 1 (S2), and the task having the task number 1 is instructed to be executed and activated (S3). After that, when a certain period of time elapses, the task with the task number 1 is instructed to stop (S4). At this time, the task instructed to stop is forcibly stopped even if it is being executed. Therefore, until the information is being saved in a memory, register, etc. have to wait. In other words, the information being executed is saved in the memory (S5), the saved information is read out at the next activation, and the process is restarted from the state at the time of the previous execution.

After that, it is judged whether or not all the tasks have been executed (S6), and if there are any tasks remaining, the task number n is incremented (S2), and as described above, the tasks are started, stopped, and information is recorded. Repeat evacuation. After all the tasks are executed once in this way, it is judged whether or not the sequence program is completed. If not completed, the task number n is initialized again (S1), and the above process is repeated.

If the tasks are switched and sequentially executed as described above, the tasks X _{1 to} X _n will be executed in the order shown in FIG. 3B. Figure 3 (b) the shows the start instruction of task X ₁ to X _n at a, is shown an instruction to stop at b, time task X ₁ to X until stop of activation
It becomes the processing time T per _n times. In the figure, α indicates the time for saving the information in the memory after stopping the tasks X _{1 to} X _n . Furthermore, as shown by the arrow in FIG.
Each of the tasks X _{1 to} X _n will receive a one-way start / stop instruction from the system program SP.

[0006]

As is clear from the above description, the tasks X _{1 to} X _n are time-slice controlled for a fixed period of time, and one-way start / stop instructions are received from the system program. From task X ₁
When X _n is stopped, it is necessary to save the information at the time of the stop in the memory in preparation for restarting the execution next. In other words, each task X ₁ to X _n to the time to stop, there than is needed time α for saving information before execution of the next task X ₁ to X _n, the Stop Task X ₁ to X _n A delay of time α occurs before the start-up, and the execution time of the sequence program becomes longer accordingly. As a result, the amount of processing per unit time decreases, and there arises a problem that it cannot be applied to equipment that requires real-time control, for example.

The present invention has been made in view of the above circumstances, and an object of the present invention is to reduce the time required for switching tasks so that the programmable controller multiprocessor has a larger processing amount per unit time than the conventional one. It is to provide a task control system and a multitasking programmable controller.

[0008]

According to a first aspect of the present invention, when a sequence program having a plurality of tasks is processed, the system program instructs each task to start and stop, thereby switching execution tasks,
In a programmable controller that apparently processes multiple tasks in parallel, when each task is instructed to stop by the system program after startup, execution continues until the time when there is no information to be saved after the time when the stop instruction is issued. After that, the task switching request is output to the system program, and when the system program receives the task switching request from the task instructed to stop, the system program instructs to activate the next task.

According to a second aspect of the present invention, a program memory for storing a sequence program having a plurality of tasks, a system memory for storing the system program,
A central processing unit that executes the sequence program stored in the program memory according to the system program stored in the system memory, and the system program includes a start instruction unit that instructs each task of the sequence program to start, and a start instruction. When a task switching request is issued from the execution task in response to the stop instruction from the stop instruction unit, the next task is designated as the execution task. A task switching unit that activates the next task later by the activation instruction unit is provided, and when a stop instruction is given from the system program, each task continues execution until the time when there is no information to be saved after the time when the stop instruction is given. After that, a stop control unit for outputting a task switching request to the system program is provided.

According to the first and second aspects of the invention, each task outputs a task switching request to the system program after continuing execution until the time when there is no information to be saved after the system program instructs the system program to stop. When the program receives a task switching request, it gives a start instruction to the next task. Therefore, the execution task does not need to save the information for restarting the execution of the task at the time of outputting the task switching request, and the time required for saving the information in the conventional method becomes unnecessary. As a result, the time required for switching between the tasks is shortened, which leads to an improvement in the processing speed as a whole.

Here, to continue the execution until the time when there is no information to be saved, generally, when the task is forcibly stopped, the information at the time of the stop is saved in order to restart the next execution of the same task. However, it means that the execution is continued until the point where the information for restarting the execution is not needed, and the information for restarting is not needed after receiving the instruction to stop the task. After executing the process up to the eyes, the execution is stopped. In other words, a task may not immediately stop execution due to a stop instruction from the system program, but after the break point is reached, execution is stopped and a task switch request is given to the system program to move to execution of the next task. Of.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 2 shows the basic configuration of a sequencer. As is well known, the sequencer includes a central processing unit (CPU) 1 including a microprocessor, a system memory (ROM and RAM) 2 storing a system program (operating system), and a program memory (ROM storing a sequence program. Or R
AM) 3, a data memory (RAM) 4 for temporarily storing data exchanged with the control target, an input / output I / O 5 which is an interface with the control target, and a power supply (not shown). Prepare There is also provided an assisting device connecting interface 6 for connecting an assisting device for creating and modifying a system program and a sequence program, and for monitoring the operation of a sequencer.

In order to execute the sequence program by the multi-task method, the sequence program is described in task units, and the system program is described in a dedicated language (here, dedicated BASIC is used). The present invention is characterized in that time slice control is performed according to the procedure shown in FIG. Here, a task number n is assigned to each task, and the number of tasks is m. That is, when the processing of the sequence program is started, the task number n is first initialized to 0 (S1), and the task number n is incremented to 1 as in the conventional method (S2). Next, the system program instructs the task having the task number 1 to start (S3), and the task having the task number 1 is executed. After that, when a certain time has elapsed, the system program requests the task with the task number 1 to stop (S4).
The processing up to this point is the same as the conventional method. However, in the present invention, even if there is a task stop request from the system program, the task is not stopped immediately, but execution is continued until there is no information to be saved on the task side. If there is no information to be saved on the task side, the task issues a task switching request to the system program (S5), and when the system program receives the task switching request, it is determined whether all the tasks have been executed (S6). If there is a task that has not been executed, the task number n is incremented (S2) and the next task is started (S2).
3). After executing all the tasks, it is determined whether the sequence program is finished (S7). If the sequence program is not finished, the task number n is initialized again and the above process is repeated.

By the way, the information to be saved on the task side means the information necessary for restarting the execution of the task next. If you stop a task at a point where it should be a break point in this way, it is not necessary to save information in preparation for the restart of task execution, and the information exchanged between tasks is also being executed. In addition, it is possible to read and write to a predetermined place such as a memory, a register, an external storage device, etc., and eventually it becomes unnecessary to save the information when the task is stopped.

That is, as shown in FIG. 1B, when the system program SP issues a start instruction a to each of the tasks X _{1 to} X _n , the tasks X _{1 to} X _n are executed and the start instruction is issued. After a lapse of a fixed time T from a, a stop instruction b is given. At this time, the tasks X _{1 to} X _n continue execution while there is information to be saved in preparation for the restart of the next execution, and if there is no information to be saved, the tasks X _{1 to} X _n send the system program SP. And a task switching request c is generated. When the task switching request c occurs, task X
_{Since 1 to} X _n are stopped and there is no information to be saved, the system program SP only gives the instruction a to start the next task. In this way, the tasks X _{1 to} X _n are sequentially executed without saving the information each time the tasks X _{1 to} X _n are stopped. Here, the task X ₁ next task from the instruction b for stopping to X _n X ₁ to X _n
If the time until the activation instruction a of is generated is β ₁ to β _n , the time occupied by each task X _{1 to} X _n for one execution is T + β _i (i = 1, 2, ...). n). In the conventional method, since the same time is T + α, if β _i ≈α,
Although there is not much difference in the time occupied by the tasks X _{1 to} X _n for one execution, most of the time β _i is the tasks X _{1 to} X _n.
Is the time used for executing the sequence program, and the time α is the time used for saving the information. Therefore, when the method of the present invention is compared with the conventional method, the time used by the central processing unit 1 for executing the sequence program is It is longer and the processing speed is substantially increased. In other words, if the same processing as in the conventional method is performed at the same time, the load per unit time of the central processing unit 1 (system program) is reduced, and the same processing as in the conventional method is performed. For example, it becomes possible to use the central processing unit 1 having a lower processing capacity than the conventional one. Further, if the processing capacity of the central processing unit 1 is about the same as the conventional one, real-time control becomes easier to realize than the conventional method.

[0016]

As described above, according to the present invention, when each task receives a stop instruction from the system program, the task switching request is issued after the execution is continued until there is no information to be saved after receiving the stop instruction. Is output to the system program, and the system program gives a start instruction to the next task when the task switching request is received, so the execution task saves the information for restarting the execution of the task at the time when the task switching request is output. Since the time required for saving information in the conventional method is no longer necessary, the time required for switching between tasks is shortened, which leads to an improvement in overall processing speed. There is.

[Brief description of drawings]

FIG. 1 shows an embodiment of the present invention, (a) is a flowchart showing a processing procedure, and (b) is an operation explanatory view showing a task execution state.

FIG. 2 is a block diagram showing a schematic configuration of a sequencer.

FIG. 3 shows a conventional example, (a) is a flowchart showing a processing procedure, and (b) is an operation explanatory view showing a task execution state.

[Explanation of symbols]

1 central processing unit 2 system memory 3 program memory 4 data memory 5 input / output I / O 6 support device connection interface a start instruction b stop instruction c task switching request SP system program X _{1 to} X _n tasks

Claims (2)

[Claims] 1. When processing a sequence program having a plurality of tasks, a system program instructs each task to start and stop, thereby switching execution tasks, and apparently processing a plurality of tasks in parallel. In the programmable controller, when each task is instructed to stop by the system program after starting,
After execution is continued until there is no more information to save after the stop instruction is issued, a task switch request is output to the system program, and the system program starts the next task when the task switch request is received from the task instructed to stop. A multitasking control system for programmable controllers, characterized by instructing. 2. A program memory for storing a sequence program having a plurality of tasks, a system memory for storing the system program, and a central processing for executing the sequence program stored in the program memory according to the system program stored in the system memory. The system program includes a device, and a start instruction unit that instructs each task of the sequence program to start,
When a task switching request is issued from the execution task in response to the stop instruction from the stop instruction unit, the next task is changed to the execution task A task switching unit that activates the next task by the activation instructing unit after designation is provided,
Each task must have a stop control unit that outputs a task switching request to the system program after the task is instructed by the system program to continue execution until the time when there is no information to be saved after the time when the stop instruction is issued. Is a multitasking programmable controller.