JPS59163647A - Task control method - Google Patents

Abstract

PURPOSE:To perform the time division processing of plural jobs with a small operand and high efficiency and to form a multi-job system of a single CPU with high speed hardware, by validating the interruption and release given from a task within a corresponding job. CONSTITUTION:A task control system contains a job pause state control table 22 and job pause release queues 25-27. A pause instruction issuing task area 23 and a pointer 24 are provided to a task for corresponding job JC of the table 22. Then a job pause instruction is issued from the area 23, and the pointer 24 indicates the queues 25-27. At the same time, the numbers of tasks X, Y... are successively stored to the queues 25-27 by the pointer 24. Thus the time division processing is carried out for plural jobs with a small operand and high efficiency, and a multi-job system using a single CPU is obtained.

Description

[Detailed description of the invention] [Field of application of Hanamei] The present invention allocates a plurality of independent tasks (jobs) within one processing device, and executes valley tasks simultaneously in parallel. This paper relates to a method for managing tasks in a multi-job system that enables multiple jobs.

〔Mirai technology〕

In a large computer system, the processing of an operating system for system management (hereinafter referred to as O8) and user programs is divided into functional units called jobs and executed.

(These 6 jobs are usually divided into smaller tasks as the processing unit and executed. If you click online rear shift 441, the processing (job 100)
), it is divided into tasks as shown in Figure 1 and executed. That is, when an event such as the external interrupt 101 occurs, a task T1 that interrupts the seventh event is activated, and a task T2 that determines a process corresponding to the event through the processing of that task and executes it is activated. . In task T2, the above is a shared memo within the device!
JGLB is referenced or changed, the processing result is displayed on the external display device 102, and task T3 is started. Task T
3, the data of the shared memo IJ (JLB) is saved to the external auxiliary storage device +* 103.As described above, the tasks of 1 summer within the job 100 communicate with each other and synchronize.
Processing for one event at the same time.

One method for performing such processing at high speed is to provide an imaginary number processing device (hereinafter abbreviated as CPU), assign a different job to each CPU, and have each CPU independently process each job. Think about what you want to do. Since this CPU executes only a single job at each time, it is called a single-job system, and its sequence is shown in FIG. In the same figure, CPUI~3 has jobs JA, JB, and J, respectively.
Each job of C is assigned σ et al. 6, Nachi et al. is task TAI~T
It is composed of 03 etc. The task management method of the single job system is to give execution priority to each valley task,
The task with the highest execution priority (A) is selected from among the tasks that have received an execution request and are in an executable state and are executed. For this reason, an execution priority order is assigned in advance according to the processing content of the task, and the processing flow of the job tray is determined.

As the performance of hardware devices has improved, it has become possible for a single CPU to perform tasks that were previously shared by multiple CPUs. In this case, multiple single job systems as shown in Figure 2 can be integrated into one computer.
I'm thinking of incorporating it into the PU, but with this, I can't start or stop jobs on a per-job basis. In addition, conflict management of the valley a quality source between the same tasks of the same job may cause problems such as affecting tasks in other jobs. In order to avoid this, it is possible to combine the broken number of jobs and reconfigure them as one job, but this requires a great deal of work such as changing the program.

A multi-job system is a system in which a single CPU can process multiple jobs. A single O8 is provided, and each job is processed under this O8. Jobs in this case can be configured using the same concept as the single job system.

FIG. 3 shows this multi-job system, which allows jobs JA to JC, which were individually executed by a plurality of CPUs in a seven-job system, to be executed by one CPU10. Therefore, O8 is a single-CPU
It has a job execution 11tlJ control function that instructs to start and stop each job for a plurality of jobs in the job. In addition, in the case of online real-time control, each resource in the CPU (shared memory area G on the main storage device) is
LB.

The auxiliary memory uses the shared memory area BULK on the device and the input/output device RI10) in a time-sharing manner.

For this reason, a macro command 5ERV/FREE for occupying or releasing these shared resources is provided. However, when this R8ERV/FREE macro instruction is used to rewrite data on the shared memory GLB, the processing of the macro instruction requires a large number of procedures such as establishing exclusive rights to the target resource and eliminating requests from others. Since the overhead is large, the macro instruction ASUSP/AR8UM has been used to simply inhibit or cancel interrupts from other tasks, regardless of the priority of the task. Among these, the macro instruction ASU8P temporarily puts all tasks other than the issuing task into a dormant state, and the macro instruction AR8UM cancels the dormant state. Interrupts by this macro instruction ASUSP/AR8UM are disabled.

The scope of cancellation is limited to single-C for single-job systems.
There is no problem because the processing range of the PU is limited within a single job, but in a multi-job system, the execution work differs for each job, so Issuance of the macro instruction ASUSP/AR8UM extends to the Ike job and affects the execution of that task, causing a problem.For this reason, in a multi-job system, a process called rewriting the shared memo IJGLB is required. In order to execute even a small amount of processing, the macro instruction R8ERV/FR,EE, which takes a long time to execute, must be used, resulting in a drawback that a large processing overhead is unavoidable.

[−) Self-evident purpose]

An object of the present invention is to overcome the above-mentioned drawbacks of the conventional technology and to enable interrupts from tasks to be inhibited and canceled on a job-by-job basis in a multi-job system with a single-CPU and single-O8 configuration. , provides a unique task management method that significantly reduces processing overhead.

[Summary of the invention]

The present invention provides a table for managing job status, and when a task accesses a shared memory area, it issues a job suspension command to place only the job to which the task belongs on the above table in a suspended state, and while the task is accessing the shared memory area, the same job Even if an interrupt occurs from another task, if the job is in a dormant state by O8, the execution of the interrupt is made to wait until the access is completed.

[Implementation sequence of the invention]

The present invention will now be described by way of example. In this implementation sequence,
First, the job status management table (JCTB
) 21 is provided. This shows the status of each job in the CPU (
This is a table for storing information (operation/pause/stop), and depending on the contents, the following control is performed by O8, which will be described later. That is, if it is displayed as being running, a startup request for the task within the job is accepted and executed. If the task is displayed as being inactive, a start request for the task within the job will be accepted, but execution will be suppressed. Additionally, if the display indicates that the task is stopped, no activation request for the task within the job will be accepted.

The status of these 1L jobs can be shifted in the direction of the arrow in the status diagram shown in FIG. 5, and is controlled by the job control macro command. Among these, the job in the initial state is stopped, and when this job is started, it is moved to the operating state. The operation state is changed to the stopped state when the use of the job is changed or when an abnormality occurs within the job. Further, the transition from the dormant state to the stopped state is also performed by changing the use of the relevant job, etc., and these transitions are performed by issuing the respective corresponding macro instructions. The transition control between the operating state and the rest state shown in FIG. 5 is deeply related to the present invention.

The transition from the body motion state to the dormant state is performed when it is desired to temporarily suspend the job, and after a certain period of time, the job must be returned to its original operating state. For this purpose, a job suspension state control table (JCBE) shown in FIG.
X) 22 and job suspension release queue (TCBI) 2
5 to 27 are provided. The contents of this table 22 corresponding to each job are shown only for job JC in FIG. It consists of a pointer 24 pointing to the suspension release queues 25-27. In the queues 25 to 27, the numbers of tasks X, Y, .

Further, in this embodiment, task execution queues (TCB2) 29 to 31 and a pointer (QPB) 28 for pointing them are used as shown in FIG. This samurai procession 29-31
For each task priority of 0.1゜2,...
It connects and displays tasks that have been requested to start in order using pointers, and regardless of which job the task belongs to, C
Since only one PU can be installed, they will be displayed in the same queue. Therefore, the pointer 28 is also independent of jobs and consists of pointers for each task priority. For example, as shown in FIG. 7, the pointer with priority l is task X1 of job JA.
Next, task Y of job JB is displayed.

Using the tables 21, 22, pointer 28, and queues 25 to 31 shown above, O8 performs task execution control as shown in the flowchart of FIG. Of course, the O8 also has input/output control and other system management functions, but these may be conventional ones. This task execution control program is started (1) when the currently executing task ends, (2) when the currently executing task becomes idle for various resources within the CPU, and (3) when the input/output 1tlJ Issue your request,
This occurs when the input/output control is waiting for completion, or (4) when a task with a higher priority than the currently executing task is activated due to an external interrupt or the like. And when started,
First, in step 40, the task execution queues 29 to 31 indicated by the pointer 28 are searched, and the task execution queues 29 to 31 indicated by the pointer 28 are searched and
, I'm here - first come first served in priority order. VCS
Check whether each task is executable or not, and choose the first one that is executable. However, Miyuki RJJ Noh referred to here means that the task is returned to the task execution queue 29 to 31 because it is waiting for various resources to become available or for input/output control to be completed as described in the program startup conditions in Figure 8. In the case of a task, this means that its attendance condition has been resolved when step 40 is executed. Such a determination is a general function of O8, and is different from the determination of whether or not execution is possible in step 41, which will be described later. Next, for the task selected in step 40, the status of the job to which the task belongs is determined in step 41.

The details of this step 41 are shown in FIG. 9. First, in step 411, the job number to which the currently selected task belongs is retrieved from the task execution queues 29 to 31, and in step 412, the job number to which the task currently selected belongs is retrieved. Management table 2
1 and find out the status of this job. As a result, if the job is in operation, the process moves to the next step 42. If the job is suspended, in the next step 413, the job suspension state control shown in FIG. Area 23 of table 22
It is determined by examining the task, and if the task has been suspended, the process moves to the next step 42. When the process moves to step 42, the task selected A in step 40 is made executable, and the above-mentioned area 1 (k) is assigned to the task (
Further, in step 43, initial settings of registers necessary for the execution of the task (at the time of new startup) or recovery processing (at the time of restart after interruption) are performed, and subsequent control is handed over to the task.

10,000, as a result of the determination in steps 412 and 413 in Figure 9,
If the job to which the task selected in step 40 belongs is suspended, and the task is not the task that issued the job suspension macro instruction to the job, the pointer 24
The task is placed in the job suspension release queue 25-2.
7 and put the job into a standby state. Then, the process returns to step 40 to search for the next executable task.

Next, with the support of the task execution control program explained in FIGS. 8 and 9 above, rewriting of the shared memory area, etc. can be easily performed using a simple macro instruction like the conventional macro instruction AsUSP/AR8UM. Show what you can do. Reading and writing (accessing) this shared memory area becomes a problem because the contents of that area are accessed by multiple tasks within the same job, so if task TA is rewritten in a certain step, then task TB is rewritten in a certain step. If the order of access, such as reading data, is maintained, it will operate correctly, but if this order is out of order, incorrect results may occur. However, due to the priority order of tasks, this order may be disrupted by an interrupt from task TB, and it is necessary to prevent this. Therefore, when 21 females access the shared memory area, the first
As shown in Figure 0, prior to accessing task TA
issues a job suspension macro instruction, and after the access is completed, issues a job suspension M removal macro instruction. Among these, the job suspension macro instruction performs the processing shown in FIG. νi
5. First, in step 61, the status of the job to which the task TA that issued this macro instruction belongs is checked with reference to the job status management table 21, and if it is not running, the process ends without doing anything. If the job is rarely in operation, the process moves to step 62, and the job status financial table 210 changes the display to indicate that the job is inactive. Subsequently, in step 63, the number of the task TA is registered in the area 23 corresponding to the job in the job suspension state control table 22, and the next step in FIG. Ru. Suppose, therefore, that another task TB in the same job that is m ahead in rank is activated and an interrupt occurs during this one access operation. Then, as mentioned above, the task execution I inquiry program shown in FIG. The indication that the job is stopped is set in the job status management table 21. Therefore, step 41 in step 41 of FIG.
Based on the determination in step 2,413 (FIG. 9), the task TB that has issued this imprint request is placed in the job pause release queues 25 to 27 in step 414, and its execution is continued until the job is released from the pause state. Not done. Task TA can then continue accessing the shared memory area.

Therefore, the contents of the shared memory area are prevented from being erroneously rewritten by other tasks TB within the same job.

When task TA finishes accessing the shared memory area, task TA issues the job suspension release macro command shown in Figure 10.
Published by. This is done in step 6 as shown in Figure 12.
In step 4, it is confirmed that the job is currently suspended, and then in step 65, the number of the task TA in the area 23 of the job suspension state control table 220 that was registered when the job suspension command was issued is cleared, and the job suspension is released. Task TBs registered in queues 25 to 27 are in execution queue 2.
9 to 31 and waits for execution.

Note that if the processing of the macro commands ASUSP and AR8UM of the multi-job system are functionally the same as the job suspension macro command and the job suspension release macro command shown in FIGS. 11 and 12, the conventional program can be changed. You can migrate from a single-job system to a multi-job system without having to

〔Effect of the invention〕

As is clear from the above implementation sequence, according to the present invention, interrupt suppression from a certain task can be made effective only within that job, so time-sharing processing of only a few jobs can be performed. There is an effect that the process can be performed efficiently with small overhead.Also, there is an effect that a multi-job system of a single PU using high-speed hardware can be configured by using the instructions of a conventional single-job system as is.

[Brief explanation of drawings]

Fig. 1 is a diagram showing a processing sequence in online real-time control, Fig. 2 is an explanatory diagram of a single job system, Fig. 3 is an explanatory diagram of a multi-job system, and Figs. 4 and 6 are diagrams showing jobs used in the present invention. A diagram showing an example of a state management table and a job suspension release queue, FIG. 5 is a diagram showing job state transition in a multi-job system, and FIG. 7 is a CPU free queue of tasks for which activation requests have been issued. FIG. 8 is a flowchart showing the execution sequence of the task execution control OS program, FIG. 9 is a flowchart showing details of the job status determination processing section in FIG. 8, and FIG. 10 is a macro for job suspension and suspension cancellation. Diagram illustrating the usage of instructions, 1st
1 and 12 are flowcharts showing the processing contents of job suspension and suspension release macro commands. 21... Job state management table, 22... Job suspension state control table, 23... Pause command issuing task area, 24... Pointer, 25-27... Job suspension release queue, 28...・Execution queue pointer,
29-31...Task execution queue. Agent: Mr. Tadashi Akimoto Jitsugyo 1 (2) No. 20 ¥3 [21 O ¥4 Kasumi 1 5th Cheer Rod Otsu Figure 2 γ7 Figure Q Certain lO Figure C Tee Σ Koha 11 Mouth $ 12 (2)

Claims (1)

[Claims] ■U5 describes how to fill up tasks in a system with a single processing device and a single operating system that performs time-sharing processing in units of tasks. , a memory table displaying whether each job is in a stopped state, body movement, tK state, or temporary rest state;
A suspension command to change the above-mentioned running job to the dormant state, a cancellation command to change the above-mentioned suspended job to the working state, and when one of the tasks is started, the activation task. Even if the execution task 11 is called, the task to which the activated task belongs is displayed in the dormant state on the memory table, and the activated task is not the one based on the suspended command; and in that case. In addition to the task execution system Hill program provided in the operating system, which has the function of prohibiting the execution of the activated task, any task can be
When accessing a memory area that is shared with other tasks in the job to which the task belongs, the job to which the task belongs will be placed in a dormant state by issuing the above-mentioned suspension money prior to the access. Interrupts from other tasks in the job are handled by the support of the task execution control program mentioned above, and after the above access is completed, the above-mentioned release-# lie is issued to return the above-described job to the operating state. A task management method having an additional advantage of controlling access of tasks to the shared memory area.