JP5262418B2 - Task scheduling apparatus and task scheduling method - Google Patents

Description

  The present invention relates to a task scheduling apparatus and a task scheduling method for scheduling transmission of messages performed between tasks.

  In an operating system (OS) such as a real-time operating system (RTOS), a plurality of tasks are managed. In addition, messages such as data and control signals are transmitted and received between these tasks.

  The inter-task communication method for transmitting and receiving this message is performed as follows.

  Each task has a message reception queue for storing messages. Each task is preset with a priority. The tasks are executed in the order of the highest priority, and the task being executed transmits the message in the message reception queue of its own task to the destination message reception queue. At this time, the task transmits a message in a predetermined burst amount unit, and when all the messages in the message reception queue of its own task are transmitted, the task having the next highest priority transmits the message.

  FIG. 7 is a block diagram showing a configuration of an information processing apparatus that realizes the inter-task communication method. In FIG. 7, the information processing apparatus includes an OS unit 50 and an application unit 60. The OS unit 50 is realized by an OS, and includes a ready queue management unit 51, a dispatch unit 52, and a scheduling unit 53. The application unit 60 is realized by an application program and includes a plurality of tasks.

  The ready queue manager 51 manages a priority queue (hereinafter referred to as a ready queue) for queuing tasks in descending order of priority. Queuing is a process of sequentially transmitting messages in the message reception queue without waiting for the completion of the destination process.

  The dispatch unit 52 dispatches to an executable task. The dispatch is a process for assigning CPU processing capacity.

  The scheduling unit 53 registers the priority of each task notified from the application unit 60 in the ready queue via the ready queue management unit 51, and activates the task according to the registered content. In addition, the scheduling unit 53 instructs the dispatch unit 52 to dispatch the started task.

  In such an inter-task communication method, if a task with high priority does not transmit all messages in the message reception queue of its own task, other tasks cannot transmit messages.

  For this reason, when congestion occurs in the message reception queue, there arises a problem that the message reception queue cannot store messages. For example, when a large number of messages occur suddenly, if more messages are sent than can be stored in the message reception queue of the destination task, the message reception queue can store the messages. Disappear. Messages that could not be stored are discarded.

  Such a problem occurs not only in queuing performed between tasks but also in general queuing. For example, it also occurs in the queue control device described in Patent Document 1.

  In this queue control device, a module is associated with each transmission / reception port. Each module includes a reception-side priority control queue unit, a transmission destination port queue unit, and a transmission-side queue unit.

  The reception-side priority control queue unit stores data received by a port corresponding to its own module. The transmission destination port queue unit transmits the data stored in the reception-side priority control queue unit to the transmission-side queue unit of the port that transmits the data. The transmission side queue unit temporarily stores the data received from the transmission destination port queue unit, and transmits the stored message to the transmission destination device.

  Here, if more data than the amount stored in the transmission side queue unit is transmitted from the transmission destination port queue unit, the transmission side queue unit cannot store the transmitted data.

  In this queue control apparatus, in order to solve this problem, when the amount of data stored in the transmission side queue unit exceeds a predetermined first value, data transmission by the transmission destination port queue unit is performed. To stop. In addition, when the amount of data stored in the transmission side queue unit falls below a predetermined second value, the transmission side queue unit resumes data transmission by the transmission destination port queue unit.

This makes it possible for the sending queue unit to prevent new data from being sent before it becomes impossible to store data, so that the message reception queue cannot store messages. Can be solved.
Japanese Patent Laid-Open No. 2001-244981

  In the queue control device described in Patent Literature 1, the transmission destination port queue unit and the transmission side queue unit can transmit data independently of each other. However, in inter-task communication in which messages are sent and received between tasks, as described above, if a task with a high priority does not send all the messages in its own message reception queue, other tasks send messages. I couldn't. For this reason, when the invention described in Patent Document 1 is applied to the inter-task communication method, the following problems occur.

  In other words, in inter-task communication, if the message reception queue of the destination task exceeds the first value, the destination task will be sent from the source task even if the source task stops sending messages. Not executed due to low priority. Therefore, since the task at the transmission destination cannot transmit the message in the message reception queue of its own task, the message reception queue of the task at the transmission destination does not fall below the second value.

  For this reason, when messages are transmitted and received between tasks, the message reception queue cannot store messages, and the problem that the messages are discarded cannot be solved.

  An object of the present invention is to provide a task scheduling apparatus and a task scheduling method that solve the above-described problem that a message is discarded when a message is transmitted and received between tasks.

  A task scheduling apparatus according to the present invention is a task scheduling apparatus that executes a plurality of tasks in order from a task with a higher priority, and stores priority identification means for storing identification information of each task in association with the priority of the task. A plurality of message storage means corresponding to each task and storing messages for the task, and a task identified by the identification information corresponding to the highest priority among the identification information in the priority storage means. Control means for executing, transfer means for sending a message in the message storage means corresponding to the task being executed to the message storage means of the destination of the message, and the amount of messages in the message storage means of the destination Determining means for determining whether or not a predetermined first threshold value or more; and determining the amount of the message by the determining means When it is determined that the value is greater than or equal to the value, the transmission of the message by the transfer unit is stopped, and the priority of the task corresponding to the message storage unit of the transmission destination in the priority storage unit is executed. Management means for changing the priority to a priority higher than that of the middle task and then terminating the task being executed.

  The task scheduling method according to the present invention also stores a priority storage means for storing each piece of identification information of a plurality of tasks in association with the priority of the task, and stores a message for the task corresponding to each task. A scheduling method by a task scheduling device including a plurality of message storage means, and executing a task specified by the identification information corresponding to the highest priority among the identification information in the priority storage means, A message in the message storage means corresponding to the task being executed is transmitted to the message storage means of the transmission destination of the message, and the amount of messages in the message storage means of the transmission destination is a predetermined first threshold value or more And if it is determined that the amount of the message is greater than or equal to the first threshold, The priority of the task corresponding to the message storage means of the transmission destination in the priority storage means is changed to a priority higher than the priority of the task being executed, End the task.

  ADVANTAGE OF THE INVENTION According to this invention, when transmitting / receiving a message between tasks, it becomes possible to suppress that a message is discarded.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of an information processing apparatus according to an embodiment of the present invention. In FIG. 1, the information processing apparatus 100 includes a storage unit 101 and an information processing unit 102.

  The storage unit 101 includes a ready queue 111 and a message (MSG) reception queue 112.

  The ready queue 111 is an example of a priority storage unit. The ready queue 111 stores the identification information of each of the plurality of tasks in association with the priority of the task.

  There are a plurality of message reception queues 112. Each of the message reception queues 112 corresponds to one of a plurality of tasks. The message reception queue 112 stores messages for tasks corresponding to the own queue (message reception queue 112).

  The information processing unit 102 is, for example, a CPU, reads an OS and an application program from a recording medium readable by the CPU, and executes the read OS and application program. Note that the recording medium may be the storage unit 101 or may be different from the storage unit 101.

  FIG. 2 is a block diagram illustrating a configuration example of the information processing unit 102. In FIG. 2, the information processing unit 102 includes an OS unit 1 and an application unit 11.

  The OS unit 1 is an example of a control unit, and is realized by the information processing unit 102 executing the OS. The OS unit 1 includes a system call processing unit 2, a ready queue management unit 3, a scheduling unit 4, a dispatch unit 5, and a message reception queue management unit 6.

  The system call processing unit 2 receives a system call from the application unit 11. The system call includes a change system call for changing the priority of the task, a registration system call for registering a new task, a measurement system call for measuring the amount of messages stored in the message reception queue 112 of the message destination, There are a monitoring system call for measuring the amount of messages stored in the message reception queue 112 of the message transmission source, an end system call for terminating the task being executed, and the like. Hereinafter, the amount of the stored message is referred to as a staying amount.

  The ready queue management unit 3 manages the ready queue 111 in the storage unit 101. Specifically, the ready queue management unit 3 changes or deletes information in the ready queue 111 and registers identification information and priority in the ready queue 111. More specifically, when the system call processing unit 2 receives the changed system call, the ready queue management unit 3 changes the priority of the task in the ready queue 111 according to the changed system call. Further, when the system call processing unit 2 receives the registration system call, the identification information and the priority are associated with each other and registered in the ready queue 111 according to the registration system call. Further, when the system call processing unit 2 receives the end system call, the ready queue management unit 3 deletes the information in the ready queue 111 according to the end system call.

  The scheduling unit 4 starts the tasks specified by the identification information in the ready queue 111 in order from the task with the highest priority. Specifically, the scheduling unit 4 activates the task with the highest priority in the ready queue 111, and then terminates the activated task when the system call processing unit 2 receives the termination system call. Then, the scheduling unit 4 activates the task with the highest priority in the ready queue 111.

  The dispatch unit 5 dispatches tasks started by the scheduling unit 4. As a result, the OS unit 1 executes the task of the identification information associated with the highest priority among the identification information in the ready queue 111.

  The message reception queue management unit 6 monitors each staying amount in the message reception queue 112. More specifically, when the system call processing unit 2 receives the measurement system call or the monitoring system call, the message reception queue management unit 6 monitors the staying amount of the message reception queue 112 according to the measurement system call or the monitoring system call. To do.

  The application unit 11 is realized by the information processing unit 102 executing an application program.

  FIG. 3 is a block diagram illustrating a configuration example of the application unit 11. In FIG. 3, the application unit 11 includes tasks A to C. Tasks A to C all have the same configuration. Hereinafter, task B will be described as an example, and descriptions of tasks A and C will be omitted. Further, it is assumed that task B is a task being executed.

  Task B includes a message transfer unit 12, a destination task message reception queue congestion detection unit 13, a self-task message reception queue congestion recovery detection unit 14, and a scheduler management unit 15.

  The message transfer unit 12 is an example of a transfer unit. The message transfer unit 12 transmits the message in the message reception queue 112 corresponding to its own task (task B) to the message reception queue 112 that is the destination of the message.

  The destination task message reception queue congestion detection unit (hereinafter abbreviated as congestion detection unit) 13 is an example of a determination unit. The congestion detection unit 13 monitors the amount of stay in the message reception queue 112 that is the transmission destination of the message transmitted by the message transfer unit 12. Specifically, the congestion detection unit 13 periodically issues a measurement system call for measuring the staying amount of the message reception queue 112 of the transmission destination to the OS unit 1 and monitors the staying amount based on the response information. To do.

  The congestion detection unit 13 determines whether the staying amount is equal to or greater than a predetermined congestion occurrence threshold. The congestion occurrence threshold is an example of a first threshold.

  The own task message reception queue congestion recovery detection unit (hereinafter abbreviated as a recovery detection unit) 14 is an example of a determination unit. The recovery detection unit 14 measures the staying amount of the message reception queue 112 corresponding to the own task. Specifically, the recovery detection unit 14 periodically issues a monitoring system call for measuring the staying amount of the message reception queue 112 of its own task to the OS unit 1 and monitors the staying amount based on the response information. .

  The recovery detection unit 14 determines whether the staying amount that is equal to or greater than a predetermined congestion recovery threshold is less than the congestion recovery threshold. The congestion recovery threshold is an example of a second threshold and is smaller than the congestion occurrence threshold.

  When the congestion detection unit 13 determines that the staying amount is equal to or greater than the congestion occurrence threshold, the scheduler management unit 15 stops the message transmission by the message transfer unit 12. The scheduler management unit 15 issues a change system call to the OS unit 1 to change the priority of the task corresponding to the message reception queue 112 of the message transmission destination to a higher priority than the priority of the own task. Then, the priority of the task corresponding to the message reception queue 112 of the transmission destination of the message is changed to a priority higher than the priority of the own task. Then, the scheduler management unit 15 issues an end system call to the OS unit 1 and ends its own task.

  Further, when the recovery detection unit 14 determines that the staying amount has fallen below the congestion recovery threshold, the scheduler management unit 15 issues an end system call to the OS unit 1 and sends identification information of its own task from the ready queue 111. Delete and end your task.

  Next, the operation will be described.

  FIG. 4 is an explanatory diagram for explaining a basic operation of the information processing apparatus 100. Task A is a task for receiving a message from outside the information processing apparatus 100. Task C is a task that transmits a message to the outside of the information processing apparatus 100. In addition, it is assumed that the ready queue 111 is empty.

  A message received from the outside by the information processing apparatus 100 is stored in the message reception queue 112 of the task A. When the message is stored in the message reception queue 112, the scheduling unit 4 of the OS unit 1 registers the task A identification information and the priority for task A in the ready queue 111 in association with each other.

  Subsequently, the OS unit 1 executes a task corresponding to the highest priority in the ready queue 111. Specifically, the scheduling unit 4 of the OS unit 1 activates the task specified by the identification information corresponding to the highest priority in the ready queue 111 and notifies the dispatch unit 5 of the identification information. When receiving the identification information, the dispatch unit 5 dispatches to the task specified by the identification information.

  Thereby, task A is executed. The message transfer unit 12 of the task A transmits the message in the message reception queue 112 of the own task to the message reception queue 112 of the task B. When the message reception queue 112 of task B receives the message, the message is stored.

  Further, the scheduling unit 4 of the OS unit 1 associates the identification information of task B with the priority for task B in the ready queue 111 when the message is stored in the message reception queue 112 of task B. Remember.

  When the message transfer unit 12 transmits all the messages in the message reception queue 112 of task A, the message transfer unit 12 notifies the scheduler management unit 15 of a completion report indicating that the message transmission is completed. Upon receiving the completion report, the scheduler management unit 15 generates a termination system call and transmits the termination system call to the scheduling unit 4 of the OS unit 1 via the system call processing unit 2. The end system call includes identification information of the own task.

  When the scheduling unit 4 receives the termination system call, the scheduling unit 4 performs termination processing to terminate the task specified by the identification information in the termination system call. Specifically, the scheduling unit 4 notifies the ready queue management unit 3 of the end system call. When receiving the termination system call, the ready queue management unit 3 deletes the identification information in the termination system call and the priority corresponding to the identification information. The ready queue management unit 3 transmits a deletion report indicating that the identification information and the priority have been deleted to the scheduling unit 4. When receiving the deletion report, the scheduling unit 4 ends the task being executed.

  Thereby, the task A is completed. Thereafter, the OS unit 1 executes task B corresponding to the highest priority in the ready queue 111.

  The message transfer unit 12 of the task B transmits the message in the message reception queue 112 of the own task to the message reception queue 112 of the task C. When the message reception queue 112 of task C receives the message, it stores the message.

  The scheduling unit 4 of the OS unit 1 stores the identification information of the task C and the priority for the task C in the ready queue 111 in association with the message stored in the message reception queue 112 of the task C. .

  Here, the scheduling unit 4 performs scheduling so that the reception process for receiving a message from the outside of the information processing apparatus 100 is executed with a higher priority than the transmission process for transmitting a message to the outside of the information processing apparatus 100. Specifically, the scheduling unit 4 stores the identification information and the priority in the ready queue 111 so that the task A, the task B, and the task C are in order from the highest priority.

  Next, the operation of the information processing apparatus 100 when congestion occurs in the task B message reception queue 112 will be described. Specifically, when the information processing apparatus 100 is in a normal state (state α), congestion that has occurred in the message reception queue 112 of task B is detected, and the information processing apparatus 100 transitions to a congestion state (state β). After that, the operation until the congestion generated in the message reception queue 112 of task B is recovered and the information processing apparatus 100 transitions to the normal state (state γ) will be described.

  FIG. 5 is an explanatory diagram showing the respective retention amounts of the message reception queue 112 and tasks having execution rights in those states. FIG. 6 is an explanatory diagram showing information in the ready queue 111 in these states. In FIG. 5, the message reception queue 112 is abbreviated as a reception queue. In FIG. 6, the priority is highest for [0] and decreases as the number increases.

  In state α, task A is being executed. Messages are stored in the message reception queues 112 of tasks A and B, respectively. Further, the ready queue 111 stores task A identification information (task A1) and task B identification information (task B1). The priority of task A is higher than the priority of task B. Then, the message transfer unit 12 of task A transmits the message in the message reception queue 112 of task A to the message reception queue 112 of task B.

  The congestion detection unit 13 of task A periodically generates a measurement system call that measures the staying amount of the message reception queue 112 of task B, and receives the measurement system call via the system call processing unit 2. The data is transmitted to the queue management unit 6.

  When the message reception queue management unit 6 receives the measurement system call, the message reception queue management unit 6 checks the staying amount of the message reception queue 112 of task B. The message reception queue management unit 6 generates information indicating the confirmed staying amount as response information of the measurement system call, and transmits the response information to the congestion detection unit 13 of the task A via the system call processing unit 2. To do.

  When the congestion detection unit 13 receives the response information, the congestion detection unit 13 determines whether the staying amount indicated by the response information is equal to or greater than the congestion occurrence threshold. The congestion detection unit 13 ends the operation when the staying amount is less than the congestion occurrence threshold. On the other hand, the congestion detection unit 13 determines that congestion has occurred in the message reception queue 112 of the task B when the staying amount is equal to or greater than the congestion occurrence threshold. As a result, the information processing apparatus 100 transitions to the congestion detection state.

  In the congestion detection state, the congestion detection unit 13 notifies the scheduler management unit 15 that the staying amount is equal to or greater than the congestion occurrence threshold. When the scheduler manager 15 receives the message, the scheduler manager 15 instructs the message transfer unit 12 to stop the message transmission, and stops the message transfer by the message transfer unit 12.

  The scheduler management unit 15 generates a changed system call for changing the priority of the task B, and notifies the ready queue management unit 3 of the changed system call via the system call processing unit 2 and the scheduling unit 4. The change system call includes the priority of the task B higher than the priority of the task A and the identification information (task B1) of the task B.

  When the ready queue manager 3 receives the changed system call, the ready queue manager 3 changes the priority of the task B corresponding to the identification information included in the changed system call in the ready queue 111 to the priority included in the changed system call. To do. As a result, the priority of task B is higher than that of task A. Thereafter, the ready queue management unit 3 generates information indicating that the priority change has been completed as response information of the changed system call, and the response information is transmitted via the scheduling unit 4 and the system call processing unit 2. It is transmitted to the scheduler manager 15 of task A.

  Upon receiving the response information, the scheduler management unit 15 generates a registration system call for registering the task A, and sends the registration system call to the ready queue management unit 3 via the system call processing unit 2 and the scheduling unit 4. Notice. The registered system call includes the priority for task A and task A identification information (task A2).

  When receiving the registration system call, the ready queue management unit 3 associates the priority and identification information in the registration system call and registers them in the ready queue 111. This is because the message remaining in the message reception queue 112 of task A is transmitted later.

  Thereafter, the ready queue management unit 3 generates information indicating that the registration of the task A is completed as response information of the registration system call, and the response information is transmitted via the scheduling unit 4 and the system call processing unit 2. It is transmitted to the scheduler manager 15 of task A.

  Upon receiving the response information, the scheduler management unit 15 generates a termination system call that terminates the task A, and notifies the scheduling unit 4 of the termination system call via the system call processing unit 2. The end system call includes task A identification information (task A1).

  When the scheduling unit 4 receives the termination system call, the scheduling unit 4 performs termination processing to terminate the task specified by the identification information in the termination system call. Thereby, the task A is completed.

  The OS unit 1 executes a task specified by the identification information corresponding to the highest priority in the ready queue 111. Now, since the priority of task B is higher than the priority of task A, task B is executed. As a result, the information processing apparatus 100 transitions to the state β.

  In the state β, the message transfer unit 12 of task B transmits the message in the message reception queue 112 of task B to the message reception queue 112 of task C.

  In addition, the task B recovery detection unit 14 periodically generates a monitoring system call for measuring the staying amount of the task B message reception queue 112, and sends the monitoring system call via the system call processing unit 2. The message is sent to the message reception queue management unit 6.

  When the message reception queue management unit 6 receives the monitoring system call, the message reception queue management unit 6 checks the staying amount of the message reception queue 112 of task B. The message reception queue management unit 6 generates information indicating the confirmed staying amount as response information of the monitoring system call, and transmits the response information to the recovery detection unit 14 of the task A via the system call processing unit 2. To do.

  Upon receiving the response information, the recovery detection unit 14 determines whether or not the staying amount in the message reception queue 112 of task B is below the congestion recovery threshold based on the response information. For example, the recovery detection unit 14 holds the staying amount indicated by the previous response information, the staying amount being held is equal to or greater than the congestion recovery threshold, and is less than the congestion recovery threshold indicated by the newly received response information. If there is, it is determined that the staying amount of the message reception queue 112 of task B is below the congestion recovery threshold.

  When the recovery detection unit 14 determines that the amount of stay in the message reception queue 112 is not less than the congestion recovery threshold, the operation ends. On the other hand, when the recovery detection unit 14 determines that the amount of stay in the message reception queue 112 is below the congestion recovery threshold, the recovery detection unit 14 determines that the congestion of the message reception queue 112 of task B has been recovered. Thereby, it is determined that the information processing apparatus 100 is in the congestion recovery state.

  In the congestion recovery state, the recovery detection unit 14 notifies the scheduler management unit 15 that the staying amount has fallen below the congestion recovery threshold. When the scheduler manager 15 receives the message, the scheduler manager 15 instructs the message transfer unit 12 to stop the message transmission, and stops the message transfer by the message transfer unit 12.

  The scheduler management unit 15 generates a registration system call for registering task B, and notifies the ready queue management unit 3 of the registration system call via the system call processing unit 2 and the scheduling unit 4. The registered system call includes the priority for task B and the identification information (task B2) of task B.

  When receiving the registration system call, the ready queue management unit 3 associates the priority and identification information in the registration system call and registers them in the ready queue 111. Thereafter, the ready queue management unit 3 generates information indicating that the registration of the task B is completed as response information of the registration system call, and the response information is transmitted via the scheduling unit 4 and the system call processing unit 2. It is transmitted to the scheduler manager 15 of task B.

  When the scheduler management unit 15 receives the response information, the scheduler management unit 15 generates a termination system call that terminates the task B, and notifies the scheduling unit 4 of the termination system call via the system call processing unit 2. The end system call includes task B identification information (task B1).

  When the scheduling unit 4 receives the termination system call, the scheduling unit 4 performs termination processing to terminate the task specified by the identification information in the termination system call. Thereby, the task B is completed.

  Thereafter, the OS unit 1 executes a task specified by the identification information corresponding to the highest priority in the ready queue 111. Since task A has the highest priority now, task A is executed. As a result, the information processing apparatus 100 transitions to the state γ.

  In the state γ, the message transfer unit 12 of task A transmits the message in the message reception queue 112 of task A to the message reception queue 112 of task A.

  Next, the effect will be described.

  In the present embodiment, the message transfer unit 12 transmits a message in the message reception queue 112 corresponding to the task being executed to the message reception queue 112 as a transmission destination. The congestion detection unit 13 determines whether or not the staying amount of the message reception queue 112 of the transmission destination is equal to or greater than a congestion occurrence threshold. When the congestion detection unit 13 determines that the staying amount is equal to or greater than the congestion occurrence threshold, the scheduler management unit 15 stops the message transmission by the message transfer unit 12, and the transmission destination in the ready queue 111 The priority of the task corresponding to the message reception queue 112 is changed to a priority higher than the priority of the task being executed. Then, the scheduler management unit 15 ends the task being executed.

  In this case, when the retention amount of the message reception queue 112 of the transmission destination becomes equal to or more than the congestion occurrence threshold, the priority of the task corresponding to the message reception queue 112 of the transmission destination becomes higher than the priority of the task being executed. Be changed. In addition, the task being executed is terminated.

  Therefore, it becomes possible to execute a task corresponding to the message reception queue 112 of the transmission destination before the message reception queue 112 of the transmission destination can no longer store the message. Therefore, since it becomes possible to suppress that the message reception queue 112 cannot accumulate | store a message, it becomes possible to suppress that a message is discarded.

  Further, in the present embodiment, the recovery detection unit 14 determines whether or not the staying amount in the message reception queue 112 corresponding to the task being executed is below a congestion recovery threshold value that is smaller than the congestion occurrence threshold value. If the recovery detection unit 14 determines that the staying amount has fallen below the congestion recovery threshold, the scheduler management unit 15 changes the priority, deletes the task identification information from the ready queue, and terminates the task being executed. .

  In this case, when the congestion of the message reception queue 112 is recovered, the task can be executed according to the original priority.

  In each embodiment described above, the illustrated configuration is merely an example, and the present invention is not limited to the configuration.

  For example, the tasks A to C are realized by an application program, but may be realized by middleware, firmware, a driver, or the like.

It is the block diagram which showed the structure of the information processing apparatus of one Embodiment of this invention. It is the block diagram which showed the structural example of the information processing part. It is the block diagram which showed the structural example of the application part. It is explanatory drawing for demonstrating the basic operation | movement of information processing apparatus. It is explanatory drawing which showed the retention amount of the message reception queue in each state. It is explanatory drawing which showed the information in the ready queue in each state. It is the block diagram which showed the structure of the information processing apparatus of related technology.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 OS part 2 System call processing part 3 Ready queue management part 4 Scheduling part 5 Dispatch part 6 Message reception queue management part 11 Application part 12 Message transfer part 13 Destination task message reception queue congestion detection part 14 Self task message reception queue congestion recovery Detection unit 100 Information processing apparatus 101 Storage unit 102 Information processing unit 111 Ready queue 112 Message reception queue

Claims (4)

A task scheduling device that executes a plurality of tasks in order from a task with a higher priority,
The plurality of tasks;
Priority storage means for storing identification information of each task in association with the priority of the task;
A plurality of message storage means corresponding to each task and storing messages for the task;
Control means for executing a task specified by the identification information corresponding to the highest priority among the identification information in the priority storage means ,
Each task is
When the own task is being executed, a transfer means for sending a message in the message storage means corresponding to the executing task that is the own task to the message storage means that is the destination of the message;
A determination means for determining whether or not the amount of messages in the message storage means of the transmission destination is equal to or more than a predetermined first threshold when the own task is being executed ;
The task scheduling apparatus stops transmission of the message by the transfer unit when the determination unit determines that the amount of the message is equal to or greater than the first threshold, and stores the message in the priority storage unit Management means for changing the priority of the task corresponding to the message storage means of the transmission destination to a priority higher than the priority of the task being executed, and ending the task being executed apparatus. The task scheduling apparatus according to claim 1,
Each task determines whether or not the amount of messages in the message storage means corresponding to the task that is being executed is lower than a second threshold value that is smaller than the first threshold value , when the task is being executed. Including determination means,
When the determination unit determines that the amount of the message is lower than the second threshold, the management unit deletes the identification information of the task whose priority has been changed from the priority storage unit, and executes the execution A task scheduling device that terminates a task in progress. The respective identification information of a plurality of tasks, the priority storage means for storing in association with the priority of the task corresponding to each task, and a plurality of message storage for storing the message for the task, wherein the plurality of A task scheduling method including a task , and a scheduling method comprising:
Of the identification information in the priority storage means, execute the task specified by the identification information corresponding to the highest priority,
When each task is being executed, the message in the message storage means corresponding to the task being executed that is the own task is transmitted to the message storage means that is the destination of the message,
When each task is being executed, it is determined whether or not the amount of messages in the destination message storage means is greater than or equal to a predetermined first threshold;
If it is determined that the amount of the message is equal to or greater than the first threshold, the priority of a task corresponding to the destination message storage unit in the priority storage unit is stopped. To a higher priority than the priority of the task being executed,
A task scheduling method for ending the task being executed. The task scheduling method according to claim 3, wherein
If each task is running , determine whether the amount of messages in the message storage means corresponding to the running task that is its own task is below a second threshold value that is smaller than the first threshold value,
If it is determined that the amount of the message has fallen below the second threshold, the task identification information that has changed the priority is deleted from the priority storage means,
A task scheduling method for ending the task being executed.

Images