Execution and real-time implementation of a temporary overrun scheduler

Patent metadata

Field	Value
Publication number	US-9697020-B2
Application number	US-201414319611-A
Country	US
Kind code	B2
Filing date	Jun 30, 2014
Priority date	Mar 1, 2005
Publication date	Jul 4, 2017
Grant date	Jul 4, 2017

Abstract

The automatic generation of a real-time scheduler for scheduling the execution of tasks on a real-time system is disclosed. The scheduler may allow task overruns in the execution of the tasks on the real-time system. The task overruns may occur when the execution of a task for a current sample hit is not completed before a next sample hit. When the task overruns occur, the scheduler may delay the execution of the task for the next sample hit until the execution of the task for the current sample hit is completed. The execution of the task for the next sample hit is performed after the execution of the task for the current sample hit is completed.

First claim

What is claimed is:

1. A method comprising:
performing a first execution of a model,
the performing the first execution of the model being performed by one or more devices;

obtaining information associated with occurrence of task overruns during the first execution of the model,
the obtaining the information associated with the occurrence of the task overruns being performed by the one or more devices,
the task overruns including a task overrun,
the information associated with the occurrence of the task overruns including information regarding a type of the task overrun, and
the task overrun occurring when a first task, associated with the model, does not complete within a time constraint; and

performing, during a second execution of the model, the first task based on the information associated with the occurrence of the task overruns,
the first task being performed without enforcing the time constraint to allow the task overrun that occurred during the first execution of the model to occur during the second execution of the model, and
the performing the first task being performed by the one or more devices.

2. The method of claim 1 , where the information associated with the occurrence of the task overruns further includes information regarding an acceptable quantity of overruns.

3. The method of claim 2 , further comprising:
determining a quantity of overruns that occurred during the second execution of the model after performing the first task,
determining that the quantity of overruns is less than the acceptable quantity of overruns, and
executing the model in real-time based on determining that the quantity of overruns is less than the acceptable quantity of overruns.

4. The method of claim 1 ,
where the information associated with the occurrence of the task overruns includes:
first information regarding first overruns of a first type, and
second information regarding second overruns of a second type that is different from the first type,

where the first type is the type of the task overrun,
where the first overruns include the task overrun, and
where the first information includes the information regarding the type of the task overrun.

5. The method of claim 1 , where performing the first task comprises:
performing, based on the information associated with the occurrence of the task overruns and during the second execution of the model, the first task by using task preemption.

6. The method of claim 5 ,
where the task preemption includes a lower-priority task being pre-empted by a higher-priority task during the second execution of the model, and
where the first task includes the higher-priority task.

7. The method of claim 1 , further comprising:
recording an order of execution of two or more tasks over a period of time when the second execution of the model is occurring,
the two or more tasks including the first task.

8. The method of claim 7 , where recording the order of the execution of the two or more tasks includes:
recording a first time at which the first task starts to execute; and
recording a second time at which the first task completes, and
determining the order of the execution of the two or more tasks based on the first time and the second time.

9. The method of claim 7 , where the execution of the two or more tasks includes the occurrence of the task overruns.

10. A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions that, when executed by at least one processor, cause the at least one processor to:
perform a first execution of a model;
obtain information associated with task overruns during the first execution of the model,
the task overruns including an overrun that occurs when a task, associated with the model, does not complete within a time constraint, and
the information associated with the task overruns including information regarding a type of the overrun; and

perform, during a second execution of the model, the task based on the information associated with the task overruns,
the task being performed without enforcing the time constraint to allow the task overrun that occurred during the first execution of the model to occur during the second execution of the model.

11. The non-transitory computer-readable medium of claim 10 ,
where the information associated with the task overruns includes:
first information regarding first overruns of a first type, and
second information regarding second overruns of a second type that is different from the first type,

where the first type is the type of the overrun,
where the first overruns include the overrun, and
where the first information includes the information regarding the type of the overrun.

12. The non-transitory computer-readable medium of claim 11 ,
where the first information includes information identifying a first quantity of the first overruns, and
where the second information includes information identifying a second quantity of the second overruns.

13. The non-transitory computer-readable medium of claim 10 , where the instructions further comprise:
one or more instructions that, when executed by the at least one processor, cause the at least one processor to:
determine a quantity of overruns that occurred during the second execution of the model after performing the task during the second execution of the model, and
execute the model based on the quantity of overruns and the information associated with the task overruns.

14. The non-transitory computer-readable medium of claim 10 , where the instructions further comprise:
one or more instructions that, when executed by the at least one processor, cause the at least one processor to:
perform, during the second execution of the model and based on the information associated with the task overruns, two or more other tasks, associated with the model, by using task preemption.

15. The non-transitory computer-readable medium of claim 14 , where the task preemption includes a lower-priority task, of the two or more other tasks, being pre-empted by a higher-priority task, of the two or more other tasks, during the second execution of the model.

16. The non-transitory computer-readable medium of claim 10 , where the instructions comprise:
one or more instructions that, when executed by the at least one processor, cause the at least one processor to:
record an order of execution of two or more tasks associated with the second execution of the model,
the two or more tasks including the task.

17. A system comprising:
a memory; and
one or more processors to:
perform a first execution of a model,
the first execution of the model including an execution of the model without enforcing time constraints;

obtain information associated with overruns during the first execution of the model,
the overruns including an overrun that occurs when a task, associated with the model, does not complete within a time constraint of the time constraints, and
the information associated with the overruns including information regarding a type of the overrun; and

perform, during a second execution of the model, the task based on the information associated with the overruns,
the task being performed without enforcing the time constraint to allow the overrun that occurred during the first execution of the model to occur during the second execution of the model.

18. The system of claim 17 , where, when performing the task, the one or more processors are to:
allow, based on the inform

Assignees

• Mathworks Inc

Inventors

• Maclay David

Classifications

• G06F9/455 Primary

  Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines · CPC title

• G06F9/4887 Primary

  involving deadlines, e.g. rate based, periodic · CPC title