JP7665495B2 - Process Schedule Method - Google Patents

Description

The present invention relates to a process scheduling method, and in particular to a process scheduling method that extends OS-specific process scheduling in a real-time OS, allowing the user to define the events that will trigger the start of a process and the content of the process to be started.

Conventionally, in order to guarantee the real-time nature of control application software (hereinafter, application software is also simply referred to as "application") on a controller in a control system, it was common to adopt a dedicated RTOS (Real Time Operating System) to guarantee the real-time nature of the application. However, an RTOS only has functions specialized for guaranteeing real-time nature, and cannot implement and operate technologies such as AI for data processing and connectivity with other devices, which are required in the recent IoT (Internet of Things) era. For this reason, controllers that adopt general-purpose OS such as Linux (registered trademark) are becoming widespread. When operating conventional control applications on a general-purpose OS, the order control of the application is influenced by the schedule characteristics of the general-purpose OS. In particular, when an abnormality occurs in an application, in conventional RTOS, abnormality countermeasure processing is often performed as an extension of RTOS processing. As a result, abnormality processing is processed preferentially and immediately, and the safety of the system can be maintained by abnormality countermeasures such as abnormal system shutdown. However, it is difficult to realize the same processing on a general-purpose OS. One solution to this problem is to provide a general-purpose OS with an extended scheduler that realizes the scheduler characteristics of a conventional RTOS.

In relation to this, for example, Patent Document 1 discloses a technique for scheduling a guest OS executed on a virtual computer. Patent Document 1 discloses a virtual computer system in which a guest OS scheduler is provided in a virtual computer manager for scheduling the virtual computer, and the guest OS is scheduled so as to adjust the CPU allocation to the guest OS based on the measured operating frequency of the guest OS. Patent Document 2, for example, discloses a technique for scheduling thread processing for executing VoIP. The call processing device of Patent Document 2 includes, in addition to a first scheduler that queues call processing events, a second scheduler that generates an event processing thread and extracts and executes call processing events from each event queue management object.

JP 2009-110404 A JP 2010-73214 A

Patent document 1 describes a technology that provides a new scheduler on a virtual machine manager to schedule guest OSs, while the call processing device described in patent document 2 provides a second scheduler to readjust the order of tasks to be executed within the processing deadline, preventing a decrease in CPU performance.

However, with these enhanced schedulers, the time when the scheduler runs and the schedule contents cannot be changed by the user.

Regarding application scheduling methods in a general-purpose OS, there is a demand for the user to be able to easily change the scheduling time and schedule contents. This will make it possible to implement a scheduler for control applications that was previously executed on an RTOS on a general-purpose OS. In addition, because the user can easily change the processing required before all applications are run, a variety of application scheduling methods can be realized.

The objective of the present invention is to provide a process scheduling method that enables a control application scheduler to be implemented on a general-purpose OS and allows the user to change the scheduling time and schedule contents.

The process scheduling method of the present invention is a process scheduling method for an information processing device that schedules a process executed on a general-purpose OS, and includes a step in which the information processing device wakes up an extended schedule process process that schedules a process separately from the OS when the OS executes a specified OS internal processing function, and a step in which the information processing device schedules a process according to an extended schedule process content program by the extended schedule process process.

The present invention makes it possible to implement a control application scheduler on a general-purpose OS, and provides a process scheduling method that allows the user to change the scheduling time and schedule contents.

1 is a functional configuration diagram of an information processing apparatus to which a process scheduling method according to a first embodiment is applied; 1 is a diagram illustrating a hardware/software configuration of an information processing apparatus to which a process scheduling method according to a first embodiment is applied. FIG. 13 illustrates an example of an extended schedule management table. FIG. 13 illustrates an example of an extended schedule event hook process template. FIG. 13 is a diagram illustrating an example of an extended schedule event information table according to the first embodiment. FIG. 13 is a diagram illustrating an overview of the relationship between an OS schedule and an extended schedule. 13 is a flowchart showing an extended schedule event selection process. 13 is a flowchart showing an extended schedule event hook process creating process. 13 is a flowchart showing an extended schedule event hook process. 13 is a flowchart showing an extended schedule processing content program registration process. 10 is a flowchart showing an extended schedule wake-up process according to the first embodiment. 10 is a flowchart illustrating an extended schedule process according to the first embodiment. 13 is a diagram illustrating a user interface of an OS internal processing function CallGraph screen. FIG. FIG. 11 is a functional configuration diagram of an information processing apparatus to which a process scheduling method according to a second embodiment is applied. FIG. 11 is a hardware/software configuration diagram of an information processing apparatus to which a process scheduling method according to a second embodiment is applied. FIG. 11 is a diagram illustrating an example of an extended schedule event information table according to the second embodiment. 10 is a flowchart showing an extended schedule wake-up process according to a second embodiment. 13 is a flowchart showing an extended schedule process according to the second embodiment (part 1); 13 is a flowchart showing an extended schedule process according to the second embodiment (part 2);

Each embodiment of the present invention will be described below with reference to Figures 1 to 18B.

[Embodiment 1]
Hereinafter, a first embodiment of the present invention will be described with reference to FIGS. 1 to 13. FIG.

First, the configuration of an information processing device to which the process scheduling method according to the first embodiment is applied will be described with reference to Figures 1 and 2.

As shown in FIG. 1, an information processing device 100 to which the process scheduling method of the present invention is applied has, as its functional configuration, an extended schedule event selection unit 101, an extended schedule event hook processing creation unit 110, an extended schedule processing content program registration unit 120, an extended schedule processing content program 121, an extended schedule processing unit 130, an application 140, an OS (Operating System) 150, and a memory unit 160.

OS 150 is system software that acts as an intermediary between the hardware and software that governs the operation (operation, management, and running) of the computer, and is, for example, a general-purpose OS such as Linux (registered trademark) or Windows (registered trademark).

The OS 150 of this embodiment has as its functions a schedule unit 151, an extended schedule event hook processing program 152, and an extended schedule process wake-up unit 153, which will be described later.

Here, the extended schedule is process scheduling that allows the user to define and execute new process order and contents separately from the original process schedule of OS 150.

The scheduler 151 is a functional unit that schedules OS-specific processes.

App 140 is an application software program that runs on OS 150 and performs productive roles and control. There are usually multiple apps 140, and they are executed in process units managed by OS 150.

The extended schedule event selection unit 101 is a functional unit that selects an internal OS processing function of the kernel of the OS 150 to be specified to wake up the extended schedule process unit 130 described below.

The extended schedule event hook processing creation unit 110 is a functional unit that creates the extended schedule event hook processing program 152 described below.

The extended schedule event hook processing program 152 is a program created based on the extended schedule event hook processing template 201 described later, and is executed as an event hook (start) for executing an extended schedule when an OS process by an internal function selected as an event for waking up the extended schedule process unit 130 described later is executed on the OS 150. The extended schedule event hook processing program 152 is inserted as an internal process of the OS 150, and the mechanism for the insertion may use a function provided by the OS 150, such as Ftrace in the case of Linux (registered trademark). The extended schedule event hook processing program 152 inserted into the OS 150 calls the extended schedule process wake-up unit 153 described later, and stores parameters required for the processing of the extended schedule process unit 130 in the extended schedule event information table 202 described later.

The extended schedule processing content program registration unit 120 is a functional unit that registers the extended schedule processing content program 121 described below.

The extended schedule processing content program 121 is a program that realizes the functions of an extended schedule process. This extended schedule processing content program 121 is created by the user, and is a program that realizes the processing to be executed as an extension of the execution of the internal processing function of the kernel of the OS 150 selected by the extended schedule event selection unit 101. In addition, there may usually be multiple extended schedule processing content programs 121, and they are registered in the extended schedule management table 200 described below by the extended schedule processing content program registration unit 120 based on the user's settings.

The extended schedule process wake-up unit 153 is a functional unit that wakes up the extended schedule process unit 130. The extended schedule process wake-up unit 153 is loaded into the OS 150 as an internal module of the OS 150 when the extended schedule process unit 130 is initialized. The extended schedule process wake-up unit 153 may be loaded using a function provided by the OS 150.

The extended schedule processing unit 130 is a functional unit that executes the extended schedule processing content program 121 registered in the extended schedule management table 200, described below, using parameters set in the extended schedule event information table 202, described below.

The storage unit 160 is a functional unit that stores data for implementing the process scheduling method.
The storage unit 160 holds an extended schedule management table 200 , an extended schedule event hook process template 201 , and an extended schedule event information table 202 .

Details about each table will be explained later.

Next, the hardware and software configuration of an information processing apparatus to which the process scheduling method of the present invention is applied will be described with reference to FIG.
An information processing apparatus 100 to which the process scheduling method of the present invention is applied, as shown in FIG. 1, has a hardware configuration realized by a general information processing apparatus such as a personal computer shown in FIG.

The information processing device 100 includes a CPU (Central Processing Unit) 302, a main memory device 304, a network I/F (Interface) 306, a display I/F 308, an input/output I/F 310, and an auxiliary memory I/F 312, all connected via a bus.

The CPU 302 controls each part of the information processing device 100 and loads and executes necessary programs into the main memory device 304.

The main memory device 304 is typically composed of volatile memory such as RAM, and stores the programs executed by the CPU 302 and the data referenced by it.

Network I/F 306 is an interface for connecting to network 5.

The display I/F 308 is an interface for connecting a display device 320 such as an LCD (Liquid Crystal Display).

The input/output I/F 310 is an interface for connecting input/output devices. In the example of FIG. 2, a keyboard 330 and a pointing device, a mouse 332, are connected.

The auxiliary memory I/F 312 is an interface for connecting an auxiliary memory device such as a hard disk drive (HDD) 350 or a solid state drive (SSD).

The HDD 350 has a large storage capacity and stores the programs for executing this embodiment. An extended schedule event selection program 361, an extended schedule event hook process creation program 362, an extended schedule processing content program registration program 363, an extended schedule processing content program 121, and an extended schedule process program 364 are installed in the information processing device 100.

The extended schedule event selection program 361, the extended schedule event hook processing creation program 362, the extended schedule processing content program registration program 363, and the extended schedule process program 364 are programs that respectively realize the functions of the extended schedule event selection unit 101, the extended schedule event hook processing creation unit 110, the extended schedule processing content program registration unit 120, and the extended schedule process unit 130.

In addition, the information processing device 100 is installed with an app 140 as a general application program and an OS 150 as basic software.

Furthermore, OS 150 includes, as its programs, a schedule program, an extended schedule event hook processing program 152, and an extended schedule process wake-up program (omitted). The schedule program and the extended schedule process wake-up program are programs that respectively realize the functions of the schedule section 151 and the extended schedule process wake-up section 153. As already explained, the extended schedule event hook processing program 152 is embedded as a program of OS 150.

Next, the data structure used in the process scheduling method of the present invention will be explained using Figures 3 to 5.

The extended schedule management table 200 is a table that manages information related to extended schedules, and as shown in FIG. 3, has fields for an extended schedule event identifier 200a, an event hook identifier 200b, and an extended schedule processing content program identifier 200c.

The extended schedule event identifier 200a stores an identifier for uniquely identifying an event identifier related to an extended schedule defined separately from the OS. The extended schedule event identifier is created, for example, in association with an internal OS processing function of the kernel of the OS 150 selected by the extended schedule event selection unit 101.

The event hook identifier 200b stores the identifier of the extended schedule event hook processing program 152 created by the extended schedule event hook processing creation unit 110.

The extended schedule processing content program identifier 200c stores an identifier for uniquely identifying the extended schedule processing content program 121 for executing a process related to an extended schedule as an extension of the schedule section 22 of the OS 150.

The extended schedule event hook processing template 201 is a template for the extended schedule event hook processing program 152, and as shown in FIG. 4, is in the form of a function and consists of a parameter setting processing statement 201a and an extended schedule process wake-up processing call statement 201b.

The parameter setting statement 201a is a statement that sets a parameter used in the internal processing function of the kernel of OS150. In the example shown in FIG. 4, the extended schedule event identifier stored in the extended schedule management table 200 is set to PARM0 as a parameter used in the internal processing function of the kernel of OS150.

The extended schedule process wake-up processing call statement 201b is a processing statement that calls an extended schedule process wake-up function to realize the functionality of the extended schedule process wake-up unit 153. The extended schedule process wake-up function is passed parameters such as PARM0, PARM1, PARM2, .... Here, PARM0 is the parameter set in the parameter setting processing statement 201a above, and PARM1, PARM2, ... are parameters passed from the kernel of the OS 150 and used in the extended schedule event.

This template is read by the extended schedule event hook processing creation unit 110, and is used to create the extended schedule event hook processing program 152.

The extended schedule event information table 202 is a table in which information for extended schedule events is stored, and as shown in FIG. 5, it consists of fields for an extended schedule event identifier 202a and an event hook parameter 202b.

The extended schedule event identifier 202a is an identifier that uniquely identifies an extended schedule event, and is linked to the extended schedule event identifier 200a in the extended schedule management table 200 in FIG. 3. The value of the extended schedule event identifier 202a is set by the extended schedule process wake-up unit 153 to the extended schedule event identifier (PARM0 in FIG. 4) passed from the extended schedule event hook processing program 152.

The event hook parameter information 202b is passed from the extended schedule event hook processing program 152 and is a parameter (PARM1, PARM2, ... in Figure 4) passed from the kernel internal processing function of the OS 150 that is used as an extended schedule event.

Next, an outline of the process scheduling method of this embodiment will be described with reference to FIG.
FIG. 6 is a diagram showing an overview of the relationship between the OS schedule and the extended schedule.
First, in the OS 150, an internal schedule process S01 schedules and executes an OS process and a user process resulting from application execution (S01).

When a user embeds the extended schedule event hook processing program 152 in an OS internal processing function, the extended schedule event hook processing is executed by the schedule event hook processing program from the corresponding OS process (S02).

In the extended schedule event hook processing, the extended schedule process wake-up processing is performed, and a wake-up notification is sent by the extended schedule process unit (S11).

In response to this, extended schedule processing is performed by the extended schedule processing unit (S12), and a process based on the execution of the extended schedule processing content program is scheduled and executed. The extended schedule process processing is set to have a higher execution priority than processes based on all of the applications 140.

As described above, by providing an extended scheduling function that allows the user to define the contents and execution timing of a process in addition to the process scheduling specific to the OS, the user can write the extended schedule processing content program 121 for the processing that the user wants to process before all the processes by the applications 140 run as an extension of the specific OS internal processing, and set the execution process. During the specific OS internal processing, the protected mode function of the OS ensures that the processes by all the applications 140 do not run. Since the process by the extended schedule event hook processing program 152 executed as an extension of that processing is also processed inside the OS, it is guaranteed that the processes by all the applications 140 do not run. Furthermore, since the extended schedule process wake-up processing called from the process by the extended schedule event hook processing program 152 is also an internal OS process, it is guaranteed that the processes by all the applications 140 do not run. By the extended schedule process wake-up processing, parameters are set in the extended schedule event information table 202, and the process to be run next is selected by the schedule processing of the OS. The execution priority of the extended schedule process processing is set higher than that of the processes by all the applications 140. Therefore, it is possible to guarantee that all processes by the applications 140 will not be executed before the extended schedule process unit 130 finishes processing. Therefore, if an abnormality occurs in a process by one specific application 140, it is possible to execute the user-defined extended schedule processing content program 121 as a countermeasure against the abnormality before the processes by the remaining applications 140 are executed. In other words, it is possible to execute the extended scheduling of the present invention as a user-defined process as an extension of the OS 150 scheduling. If an abnormality occurs in a process by an application, an exception handling function is executed as an internal OS process in Linux (registered trademark). The user can flexibly set the content of the extended schedule processing content program to select the exception handling function as an extended schedule event and reboot the system or execute an application other than the application in which the abnormality occurred when the abnormality occurs as the extended schedule processing content program.

Next, the process schedule method of this embodiment will be described with reference to FIGS.
This is a process executed by the extended schedule event selection unit 101 .

First, the OS internal processing function CallGraph management unit 102 of the extended schedule event selection unit 101 of the information processing device 100 displays an OS internal processing function CallGraph screen (S101). The user interface of the OS internal processing function CallGraph screen will be described in detail later.

Next, the user specifies the OS internal processing function to be displayed on the OS internal processing function CallGraph screen, and inputs the hook point for executing the extended schedule (S102).

Next, the extended schedule event selection unit 101 generates an extended schedule event identifier for uniquely identifying the extended schedule and sets it in the extended schedule management table 200 (S103).

Next, the extended schedule event selection unit 101 generates an event hook identifier corresponding to the input hook point and sets it in the extended schedule management table 200 (S104).

Next, the extended schedule event hook process creating process will be described with reference to FIG.
This is a process executed after the extended schedule event selection process shown in FIG.

First, the extended schedule event hook process creation unit 110 of the information processing device 100 reads the extended schedule event hook process template 201 (S201).

Next, the extended schedule event hook process creation unit 110 reads the extended schedule event identifier from the extended schedule management table 200 into PARM0 of the extended schedule event hook program (S202).

Next, the extended schedule event hook process creation unit 110 generates code for a parameter setting process statement 201a that assigns the read extended schedule event identifier (S203).

Next, the extended schedule event hook process creation unit 110 associates the event hook identifier related to the extended schedule event identifier with the extended schedule event hook program, and inserts the extended schedule event hook program into the OS internal processing function corresponding to the input hook point (S204).

The event hook identifier and the extended schedule event hook program can be associated with each other, for example, by using the function name of the extended schedule event hook program as the event hook identifier (in the example of Figures 3 and 4, HOOKFUNC-1 to HOOKFUNC-n are generated as event hook identifiers from the template HOOKFUNC-♯). Inserting an extended schedule event hook program means embedding the program's execution code in the corresponding location of the binary code of the kernel of OS 150, like the Ftrace utility of Linux (registered trademark), for example.

Next, the extended schedule event hook process will be described with reference to FIG.
This is a process executed after the extended schedule event hook process creating process shown in FIG. 8, and is a process executed by the extended schedule event hook process content program 152.

First, the extended schedule event hook process content program 152 sets an extended schedule event identifier in PARM0 (S301). The code for the parameter setting process statement 201a for this purpose is generated in S203 of FIG. 8.

Next, the extended schedule event hook process content program 152 sets the parameters of the called OS internal processing function to PARM1, PARM2, ... in order (S302).

Next, the extended schedule event hook processing content program 152 passes the parameters PARM0, PARM1, PARM2, ... and calls the extended schedule process wake-up processing (S303).

Next, the extended schedule processing content program registration process will be described with reference to FIG.
This is a process executed by the extended schedule processing content program registration section 120 .

First, the user specifies the extended schedule processing content program 121 in the information processing device 100, and inputs an extended schedule event identifier for identifying the extended schedule event that executes the extended schedule processing content program 121 (S401).

Next, the extended schedule processing content program registration unit 120 of the information processing device 100 sets the extended schedule processing content program identifier that corresponds to the input extended schedule event identifier and uniquely identifies the extended schedule processing content program 121 to the extended schedule processing content program identifier 200c in the extended schedule management table 200 (S402). The extended schedule processing content program identifier may be, for example, an executable file name or a function pointer, as long as it can uniquely identify a program.

Next, the extended schedule wake-up process according to the first embodiment will be described with reference to FIG.
This is a program executed by the extended schedule wake-up processing unit 153, and is called by executing the extended schedule event hook program inserted into the OS internal processing function by the extended schedule event hook processing creation processing of FIG.

First, the extended schedule wake-up processing unit 153 of the information processing device 100 reads PARM0, PARM1, PARM2, ... as parameters when called (see FIG. 4) (S501).

Next, the extended schedule wake-up processing unit 153 assigns PARM0 to the extended schedule event identifier 202a in the extended schedule event information table 202, and assigns PARM1, PARM2, ... to the event hook parameters 202b (S502). PARM0 is the extended schedule event identifier assigned by the extended schedule event hook program, and PARM1, PARM2, ... are subsequent parameters passed from the OS internal processing function.

Next, the extended schedule wake-up processing unit 153 sets the extended schedule process processing to a higher priority than processes by other apps 140 in the scheduler unit 151 of the OS 150 (S503).

Next, the extended schedule wake-up processing unit 153 passes the identifier of the extended schedule event to be started to the extended schedule processing unit 130 and issues a wake-up notification (S504).

The extended schedule wake-up processing unit 153 operates as a kernel internal process, so other applications 140 will not run processes until this process is completed. The wake-up notification to the extended schedule process unit 130 is sent from the OS internal processing function to first execute the extended schedule process processing.

Furthermore, the extended scheduled process is set to run with a higher priority than processes by other applications, ensuring that the extended scheduled process is always executed before processes by other applications run.

Next, the extended schedule process according to the first embodiment will be described with reference to FIG.
This is processing executed by the extended schedule processor 130, and is executed in response to a wake-up notification being sent to the extended schedule wake-up processor 153.

First, the extended schedule process unit 130 of the information processing device 100 determines whether or not there is a wake-up notification from the extended schedule process wake-up unit 153 (S601). If there is a wake-up notification (S601: YES), the process proceeds to S602. If there is no wake-up notification (S601: NO), the process waits (S610).

When the extended schedule process wake-up unit 153 notifies the extended schedule process unit 130 to wake up, the extended schedule process unit 130 searches the extended schedule event information table 202 using the extended event identifier passed from the extended schedule process wake-up unit 153 as a key, and obtains the event hook parameters (S602).

Next, the extended schedule process unit 130 searches the extended schedule management table 200 using the extended event identifier passed from the extended schedule process wake-up unit 153 as a key, and obtains the extended schedule processing content program identifier (S603).

Next, the extended schedule process unit 130 passes the event hook parameters acquired in S602 and starts a process by the extended schedule processing content program 121 corresponding to the extended event identifier (S604).

Next, the user interface of the OS internal processing function CallGraph screen will be described with reference to FIG.
FIG. 13 is a diagram for explaining the user interface of the OS internal processing function CallGraph screen.

The OS internal processing function CallGraph screen 600 is a screen that displays the processing flow and function call relationships of OS internal processing functions executed by the kernel of OS 150, as shown in FIG. 13. The black circles 601 in the figure are points where function calls are made. The intersections of black circles 601 indicate that two or more OS internal processing functions make function calls to the same function.

When embedding hook points in OS 150 and performing extended scheduling, it is necessary to carefully select the OS kernel processes to be used as schedule events in order to prevent too many schedule events from occurring. For this reason, it is advisable for the user to use the OS Internal Processing Function CallGraph screen 600 to select an OS internal processing function that has few common parts that are called from multiple kernel processes.

The user selects an OS internal processing function executed in the kernel of OS 150 that the user wants to use as a hook point for the extended schedule event from that CallGraph. When selecting, the user selects an OS internal processing function that does not overlap with other OS internal processing functions. This makes it possible to reduce the number of times that extended schedule process processing is performed.

For this purpose, in the process of inputting a hook point for executing an extended schedule in S102 of FIG. 7, for example, the relevant OS internal processing function display 602 is specified, the mouse is right-clicked to display the context menu 603, and the "specify hook point" command is input.

At this time, the extended schedule event identifier may be automatically generated by the system from the OS internal processing function name, or may be input by the user via a dialogue (not shown), etc.

As described above, in this embodiment, by setting a hook point in an internal processing function executed by the OS kernel, defining the startup timing and processing content, and starting an extended schedule process with a higher priority than other apps, it is possible to perform extended scheduling of a new process separate from the scheduling of the OS process.

This makes it possible for the control application's scheduler to be implemented on a general-purpose OS, enabling scheduling of processes whose scheduling time and schedule contents can be changed by the user.

For example, if an abnormality occurs in a specific app, the abnormality countermeasure process can be registered as a user-defined extended schedule process, and the user-defined process can be executed before the remaining apps are run.

[Embodiment 2]
A second embodiment of the present invention will be described below with reference to FIGS. 14 to 18B.

In the first embodiment, we have described a process scheduling method that allows users to define extended schedules by adding startup timing and processing content in addition to the OS schedule.

This embodiment is a process scheduling method based on the same concept as the first embodiment, but differs in that it is based on an information processing device having multiple CPU cores as the processor.

In other words, in the information processing device 100 of the first embodiment, one CPU 300 executes the OS 150 and the application 140, but in this embodiment, the CPU 300 has multiple CPU cores, and each of them can execute a process individually. Then, corresponding to each of these CPU cores, an extended schedule process unit is provided, and each of them can execute an extended schedule separate from the OS 150.

First, the configuration of an information processing device to which the process scheduling method according to the second embodiment is applied will be described with reference to Figures 14 and 15.

The information processing device 100 to which the process scheduling method of the present invention is applied has a functional configuration that is almost the same as that of the first embodiment shown in FIG. 1, but differs in that there are multiple extended schedule process units 130 (in the figure, indicated as 130a, 130b, ...) for each CPU core 301 (described below). Then, as in the first embodiment, the processing of the extended schedule process units 130 is executed with a higher priority than the applications 140 executed by each CPU core 301.

Unlike the first embodiment, the extended schedule process wake-up unit 22 wakes up all extended schedule process units 130 running on each CPU core 301.

The memory unit 160 also has a schedule exclusive flag 210 for adjusting the scheduling of multiple extended schedule process units 130. There is only one schedule exclusive flag 210 in the system. Details of setting the schedule exclusive flag 210 will be explained later.

In addition, the format of the extended schedule event information table 202 is different (details will be described later).

The information processing device 100 to which the process scheduling method is applied has a hardware configuration that is almost the same as that of the first embodiment shown in FIG. 2, but as shown in FIG. 15, the CPU 300 has multiple CPU cores 301 (in the figure, indicated as 301a, 301b, ...) each capable of executing a process individually.

Next, a data structure used in the process scheduling method of the present invention will be described with reference to FIG.
The table used in this embodiment is different from that in the first embodiment only in the extended schedule event information table 202 .

As shown in FIG. 16, the extended schedule event information table 202 of this embodiment is composed of fields for an extended schedule event identifier 202a, an event hook parameter 202b, and an execution CPU core identifier 202c, and differs only from the extended schedule event information table 202 of embodiment 1 shown in FIG. 5 in that an execution CPU core identifier 202c is added.

The execution CPU core identifier 202c stores an identifier that uniquely identifies the CPU core 301 that executes the extended schedule process corresponding to the relevant extended schedule event identifier. This causes the specified CPU core 301 to execute the scheduled process by processing the corresponding extended schedule process unit and running the extended schedule processing content program 121.

Next, the process schedule method of this embodiment will be described with reference to FIG. 17, FIG. 18A, and FIG. 18B.
In this embodiment, only the extended schedule wake-up processing and the extended schedule process processing are different from the process scheduling method shown in the first embodiment.

First, the extended schedule wake-up process according to the second embodiment will be described with reference to FIG.
First, the extended schedule wake-up processing unit 153 of the information processing apparatus 100 reads PARM0, PARM1, PARM2, . . . as parameters when it is called (see FIG. 4) (S701).

Next, the extended schedule wake-up processing unit 153 assigns PARM0 to the extended schedule event identifier 202a in the extended schedule event information table 202, and assigns PARM1, PARM2, ... to the event hook parameters 202b (S702). PARM0 is the extended schedule event identifier assigned by the extended schedule event hook program, and PARM1, PARM2, ... are subsequent parameters passed from the OS internal processing function.

Next, the extended schedule wake-up processing unit 153 sets the extended schedule process processing to a higher priority than processes by other apps in the scheduler unit 151 of the OS 150 (S703).

Next, the extended schedule wake-up processing unit 153 obtains the CPU core identifier of the CPU core on which it is being executed by a system call, and sets this as the executing CPU core identifier 202c in the extended schedule event information table 202 (S704).

Next, the extended schedule wake-up processing unit 153 passes the extended schedule event identifier to be started to all extended schedule processing units 130 and issues a wake-up notification (S705).

Next, an extended schedule process according to the second embodiment will be described with reference to FIGS. 18A and 18B.
First, the extended schedule process unit 130 of the information processing device 100 determines whether or not there is a wake-up notification from the extended schedule process wake-up unit 153 (S801 in FIG. 18A ). If there is a wake-up notification (S801: YES), the process proceeds to S802. If there is no wake-up notification (S801: NO), the process waits (S820).

When the extended schedule process unit 130 receives a wake-up notification from the extended schedule process wake-up unit 153, the extended schedule process unit 130 reads the execution CPU core identifier from the extended schedule event information table 202 using the extended event identifier passed from the extended schedule process wake-up unit 153 as a key (S802).
Next, the extended schedule process unit 130 obtains, by a system call, the CPU core identifier of the CPU core on which the extended schedule process unit 130 is being executed (S803).

Then, the extended schedule process unit 130 compares the CPU core identifier of the CPU core on which it is being executed with the execution CPU core identifier (S804), and if they match (S804: YES), proceeds to S805, and if they do not match (S804: NO), returns to S801.

When the CPU core identifier of the CPU core on which it is being executed matches the execution CPU core identifier, the extended schedule wake-up processing unit 153 refers to the schedule exclusivity flag 210 and determines whether it is ON or OFF (S805).

When the schedule exclusive flag 210 is ON, it means that the process has been scheduled by another CPU core and that the process based on that scheduling is being executed. When the schedule exclusive flag 210 is OFF, it means that the process has not been scheduled by another CPU core and new scheduling is possible. Note that the schedule exclusive flag 210 is initialized to OFF when the system is started up.

When the schedule exclusive flag 210 is ON (S805: ON), the process waits until it is turned OFF (S830).

When the schedule exclusive flag 210 is OFF (S805: OFF), proceed to S806 in Figure 18B.

When the schedule exclusive flag 210 is OFF, the extended schedule process unit 130 sets the schedule exclusive flag 210 to ON (S806).

Next, the extended schedule process unit 130 searches the extended schedule event information table 202 using the extended event identifier passed from the extended schedule process wake-up unit 153 as a key, and obtains the event hook parameters (S807).

Next, the extended schedule process unit 130 searches the extended schedule management table 200 using the extended event identifier passed from the extended schedule process wake-up unit 153 as a key, and obtains the extended schedule processing content program identifier (S808).

Next, the extended schedule process unit 130 passes the event hook parameters acquired in S802 and starts a process based on the extended schedule processing content program corresponding to the extended event identifier (S809).

Next, the extended schedule processing unit 130 sets the schedule exclusivity flag 210 to OFF (S810).

According to this embodiment, in a multi-core system, the startup timing and processing content can be defined for each CPU, and each can execute an extended schedule separate from the OS 150.

100...information processing device, 101...extended schedule event selection unit, 102...OS internal processing function CallGraph management unit, 110...extended schedule event hook processing creation unit, 120...extended schedule processing content program registration unit, 121...extended schedule processing content program, 130...extended schedule process unit, 140...application, 150...OS (operating system), 151...schedule unit, 152...extended schedule event hook processing program, 153...extended schedule process wake-up unit, 160...storage unit, 200...extended schedule management table, 201...extended schedule event hook processing template, 202...extended schedule event information table, 210...schedule exclusive flag

Claims (5)

1. A process scheduling method for an information processing device for scheduling a process executed on a general-purpose OS, comprising:
a step of waking up an extended schedule process that schedules a process separately from the OS when the OS executes a specified internal OS processing function by the information processing device;
the information processing device includes a step of scheduling a process according to an extended schedule processing content program by the extended schedule process processing,
moreover,
11. A process scheduling method comprising the step of: selecting an OS internal processing function that initiates an extended schedule process in accordance with information on an OS internal processing function input by a user, said information processing device selecting an OS internal processing function that initiates an extended schedule process in accordance with information on an OS internal processing function input by a user. 1. A process scheduling method for an information processing device for scheduling a process executed on a general-purpose OS, comprising:
a step of waking up an extended schedule process that schedules a process separately from the OS when the OS executes a specified internal OS processing function by the information processing device;
the information processing device includes a step of scheduling a process according to an extended schedule processing content program by the extended schedule process processing,
moreover,
11. A process scheduling method comprising the step of: registering, by the information processing device, an extended schedule processing content program to be started by the extended schedule process processing, in accordance with information on the extended schedule processing content program input by a user. 1. A process scheduling method for an information processing device for scheduling a process executed on a general-purpose OS, comprising:
a step of waking up an extended schedule process that schedules a process separately from the OS when the OS executes a specified internal OS processing function by the information processing device;
the information processing device includes a step of scheduling a process according to an extended schedule processing content program by the extended schedule process processing,
A process scheduling method, comprising the steps of: embedding an extended schedule event hook processing program as code for the OS internal processing function; and waking up an extended schedule process processing by the extended schedule event hook processing program. 4. The process scheduling method according to claim 3, wherein the information processing device holds an extended schedule management table that associates an extended schedule event identifier that uniquely identifies an extended schedule event, an event hook identifier that uniquely identifies a point where the event is embedded as code for the OS internal processing function, and an extended schedule processing content program identifier that uniquely identifies the extended schedule processing content program. 1. A process scheduling method for an information processing device for scheduling a process executed on a general-purpose OS, comprising:
a step of waking up an extended schedule process that schedules a process separately from the OS when the OS executes a specified internal OS processing function by the information processing device;
the information processing device includes a step of scheduling a process according to an extended schedule processing content program by the extended schedule process processing,
The information processing device has a plurality of CPU cores each of which executes a process;
a process scheduling method for performing the extended schedule process processing exclusively for each of the CPU cores;

Images