JPH1011327A - System for monitoring abnormality in multitask - Google Patents

Abstract

PROBLEM TO BE SOLVED: To monitor whole tasks and to specify the task in an abnormal state by permitting monitor CPU to successively transmit a monitor code to the respective tasks of CPU to be monitored in priority order or reverse priority order unless a response code is returned within a fixed time and recognizing the response codes from the respective tasks. SOLUTION: The tasks are started in CPU 11 by a fixed cycle and the monitor code which designates the lowest-order priority order number N is transmitted to CPU 21. CPU 21 receives the monitor code by OS 22, starts the Z task where the priority order number is the lowest-order one N and gives the monitor code N. The Z task returns the response code with the priority order number N and Z task ID as a parameter to OS 22 and CPU 21 returns the same code to CPU 11. CPU 11 compares the priority order number N of the transmitted monitor code with the priority order number N of the parameter of the response code so as to judge that the respective tasks during an operation in OS 22 are normal at the time of coincidence.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multi-task abnormality in a control device or a communication device having two or more CPUs, in which one CPU is used for monitoring and the other CPU monitors a plurality of tasks operating. It relates to a monitoring method.

[0002]

2. Description of the Related Art There have conventionally been many examples in which a plurality of CPUs are used in a control device (or a communication device) or the like. FIG.
FIG. 3 is a diagram showing a configuration example of a control device controlled by two CPUs. As shown in the figure, the control device 10 is a host CPU.
11, OS (operating system) 12, multiple tasks 13, lower CPU 21, OS 22, multiple tasks (A task 23-1, B task 23-2, C task 2)
3-3... Z task 23-n) and other hardware (omitted in the figure).

Usually, the OS 12 and the OS 22 perform multi-task control, and often operate a plurality of tasks. FIG. 5 is a diagram illustrating an example of multitask control by the OS.
Description will be made with reference to FIG. The OS 22 has a plurality of tasks (A task 23-1, B task 23-2, C task 23-3,
Controlling the Z task 23-n) and activating the corresponding task when an event occurrence signal is received from the outside (event driven method). The tasks are assigned priorities at the time of registration (priority is fixed).

For example, when an event for starting the B task 23-2 occurs from outside (a), the OS 22
-2 is started. C with lower priority during execution
If there is an activation request (event occurrence) of the task 23-3 (b), the C task 23-3 is executed after the completion of the B task 23-2. If there is an activation request (event occurrence) of the A task 23-1 having a higher priority during the execution of the C task 23-3 (c), the C task 23-3 is temporarily suspended and the A task 2 is stopped.
3-1 is executed, and after completion, the C task 23-3 is continuously executed. The same control as described above is also performed by the OS 12 of the CPU 11, data is transmitted and received between the CPU 11 and the CPU 21, and the control device 10 operates.

As a monitoring method for monitoring whether or not each task is operating normally in the above-described control method, a conventional method uses one CPU 11 (monitoring CPU) to monitor the other CPU 21.
The monitoring code is periodically transmitted to the (monitored CPU), and the received task is actuated by the received CPU 21 to return a response code to the CPU 11. The CPU 11 determines that the CPU 21 (monitored CPU) is normal if there is a response within a certain period of time, and judges that it is abnormal if there is no response within a certain period of time, and outputs a necessary message (omitted in the figure). In the above description, the monitoring CPU and the monitored CPU may alternately monitor each other.

[0006]

However, in the conventional monitoring method, a monitoring task for processing the received monitoring code is determined. If an abnormality occurs in a task having a lower priority than the monitoring task, the monitoring is performed. Since the task responds normally, it does not become an abnormality monitor. If a task with a higher priority than the monitoring task becomes abnormal and runs away, the monitoring task does not operate, and the monitoring CPU can detect the occurrence of the abnormality because no response code is returned. There is a problem that the task in the state cannot be specified and the factor analysis cannot be performed.

The present invention has been made in view of the above points, and provides a multitask abnormality monitoring method capable of monitoring all tasks and specifying an abnormal state task in order to eliminate the above problems. The purpose is to do.

[0008]

SUMMARY OF THE INVENTION In order to solve the above problems, the present invention comprises a plurality of CPUs, a plurality of prioritized tasks operated by the CPUs, and a multitask control according to the priorities. Controlled by one CPU
Are used as monitoring CPUs, and the other CPUs are used as monitored CPUs.
1 and 2, the monitored CPU returns a response code within a predetermined time to monitor the abnormality. In the multitask abnormality monitoring method, as shown in FIGS. A means for returning a response code for each task is provided.
A monitoring code is sent to the task with the lowest priority, and the task returns a response code within a certain period of time to monitor the presence or absence of an abnormality, and a response code is returned within a certain period of time. If there is no monitoring CPU, the monitoring CPU sequentially assigns monitoring tasks to the tasks of the monitored CPU in the order of priority or in the reverse order of priority.
A task in an abnormal state is specified by transmitting a response code from each task and confirming a response code from each task.

[0009]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a diagram showing an operation example 1 of the multitask abnormality monitoring method of the present invention. CPU
11 is a diagram illustrating an example of an operation in a normal state where 11 is a monitoring CPU and CPU 21 is a monitored CPU (see FIG. 4). OS22
A tasks 23-1 and B tasks 23- controlled by
2, C task 23-3... Z task 23-n responds to the received monitoring code (see FIG. 3).
Is provided. Each task is assigned a priority number, and the tasks are treated with higher priority in numerical order (lower order). The priority of each task is A task 23-1> B task 23-2> C task 23-3 >>...> Z task 23
-N.

A description will be given with reference to FIG. (1) CPU 11
In this case, the task 13 is started at a fixed period, and a monitoring code designating the lowest priority number N is transmitted to the CPU 21. (2) In the CPU 21, the OS 22 receives the monitoring code, activates the Z task 23-n having the lowest priority number N, and passes the monitoring code N. (3) The Z task 23-n receives the response code using the priority number N and the Z task ID as parameters.
CPU 21 returns the response code to OS 22
Reply to U11 (monitoring CPU). (4) CPU 11
(Monitoring CPU) is the priority number N of the transmitted monitoring code.
And the priority number N of the parameter of the response code.
If they are the same, it is determined that each task running on the OS 22 is normal.

FIG. 2 is a diagram showing an operation example 2 of the multitask abnormality monitoring method of the present invention.
FIG. 9 is a diagram showing an operation example when an abnormality occurs in a task operating in FIG. Description will be made with reference to FIG. (1) CPU 11
In this case, the task 13 is started at a fixed period, and a monitoring code designating the lowest priority number N is transmitted to the CPU 21. (2) The CPU 21 receives a command from the OS 22 and issues a command to start the Z task 23-n having the lowest priority number N, but there is an abnormal (runaway) task among the tasks with higher priority. In this case, the Z task 23-n is not activated.
Therefore, a response code cannot be returned.

(3) After a certain period of time, if there is no reply, the CPU 11 designates the priority number N-1 and sends a monitoring code with the priority increased by one to the CPU 21. (4)
In the CPU 21, the received data is received by the OS 22 and the priority number N-1 is received.
Is issued, but if there is an abnormal (runaway) task among the tasks of higher priority, the task is not started and a response code cannot be returned. If no response code is returned within a predetermined time, the above operation is repeated while sequentially increasing the priority (in the figure, the place where the priority is raised to 3 is shown).

The above operation is sequentially repeated, and FIG.
When a monitoring code designating priority number 2 is transmitted from the PU 11, (6) the B task is started by the CPU 21 for the first time, and (7) the B task 23-2 has the priority number 2 and the ID of the B task as parameters. This is an example of responding by creating a response code as a data. (8) When the response code is received by the CPU 11, the task 13 compares the priority number 2 of the monitoring code with the priority number 2 of the response code and confirms that they are the same. C task 2 corresponding to priority number 3
It is determined that an abnormality has occurred in 3-3 and output to the outside (omitted in the figure). In the above description, the tasks are started in order from the task with the lowest priority, but may be started in order from the task with the highest priority.

In the abnormality monitoring system described above, the CPU 11 is described as a monitoring CPU and the CPU 21 is monitored CPU. However, mutual monitoring can be performed by replacing each other.

As described above, according to the present embodiment, the function of returning a response code to each of a plurality of tasks is provided, and the monitoring code transmitted from the monitoring CPU is monitored. A response code is returned from each task of the CPU.
When an error occurs, a task in an abnormal state and a task with a lower priority than the task cannot return a response code, but the monitoring CPU checks the operation of the tasks in the order of priority to identify a task in an abnormal state. Can be specified.

[0016]

As described above, according to the present invention, the following excellent effects can be obtained. (1) By providing a means for returning a response code from each task of the monitored CPU, when an error occurs, a task in an abnormal state and a task having a lower priority than the task are returned as a response code.
However, the monitoring CPU can identify the task in an abnormal state by checking the operation of the tasks in order of priority.

(2) By specifying a task in an abnormal state, it is possible to find a defect that could not be found conventionally, and it is easier to find the cause.

[Brief description of the drawings]

FIG. 1 is a diagram showing an operation example 1 of the multitask abnormality monitoring method of the present invention.

FIG. 2 is a diagram showing an operation example 2 of the multitask abnormality monitoring method of the present invention.

3A and 3B are diagrams showing examples of a monitoring code and a response code, wherein FIG. 3A shows an example of a monitoring code, and FIG. 3B shows a response code.
Here is an example.

FIG. 4 is a diagram illustrating an example of a block configuration of a control device controlled by two CPUs.

FIG. 5 is a diagram illustrating an example of multitask control by an OS.

[Explanation of symbols]

 Reference Signs List 10 control device 11 CPU 12 OS 13 task 21 CPU 22 OS 23-1 A task 23-2 B task 23-3 C task 23-n Z task

Claims (1)

[Claims] 1. A system comprising a plurality of CPUs, a plurality of prioritized tasks operated by the CPUs, and controlled by multitask control according to the priorities.
U is used as a monitoring CPU, and the other CPUs are used as monitored CPUs. The monitoring code is periodically transmitted from the monitored CPU to the monitored CPU, and the monitored CPU returns a response code within a predetermined time. In the multi-task abnormality monitoring method for monitoring the abnormality, a means for returning a response code for each task of the monitored CPU is provided, and the monitoring CPU periodically determines the lowest priority of the monitored CPU. A monitoring code is sent to the task, and the task returns a response code within a predetermined time to monitor the presence or absence of an abnormality. If the response code is not returned within a predetermined time, the monitoring CPU Sends the monitoring code to each task of the monitored CPU in the order of priority or in the reverse order of the priority, and identifies the task in the abnormal state by checking the response code from each task. To do Multitasking abnormality monitoring method characterized.