WO2015128914A1

WO2015128914A1 - Software-installed apparatus, and software updating method

Info

Publication number: WO2015128914A1
Application number: PCT/JP2014/005658
Authority: WO
Inventors: 昌行桐村
Original assignee: 三菱電機株式会社
Priority date: 2014-02-27
Filing date: 2014-11-11
Publication date: 2015-09-03
Also published as: TW201533664A; CN106030539A; WO2015128914A9; JP6042025B2; JPWO2015128914A1

Abstract

　The present invention is provided with a software-installed apparatus (1) characterized in having: a storage unit which is designed to realize the updating of software in a small storage area, and in which are stored a first program (100) for controlling a device (14) connected to the apparatus and a second program (200) smaller in size than the first program (100), the second program (200) including only some of a plurality of functions of the first program (100); and a control unit, connected to the storage unit, which, when a request for updating the first program (100) is accepted, stops executing the first program and updates the first program (100) by overwriting, and also executes some functions of the second program while updating the first program.

Description

Software-equipped equipment and software update method

The present invention relates to a software update method for software-equipped devices.

In an information device that can download and update software through communication means such as the Internet, it is required to update the software simultaneously while continuing the operation of the device.

In response to such a request, a method has been proposed in which an area for storing all update programs including a correction program is secured in advance.

In this method, the update program is downloaded to the reserved area in advance while the current program is being executed. After the download is completed, the current program is replaced with the update program.

In Patent Document 1, in an environment where a plurality of OSs (Operating Systems) operate on the same H / W, a time during which processes with the same processing content executed when updating the software of each OS is executed in parallel There has been proposed a method of adjusting the execution order so that is shortened.

Patent Document 2 proposes a method of storing a means for determining whether a device is used or not, and automatically downloading software from a server when the device is not used.

In Patent Document 3, a normal state in which an application can be executed in a normal energized state and a standby state in which the power consumption is limited and the application is not executed can be switched, and the application is downloaded in the standby state. Has been proposed.

JP 2012-181578 A JP 2012-018657 A JP 2013-099557 A

In conventional software-equipped devices, it is necessary to prepare a free storage area equivalent to the program to be updated. In a conventional software-equipped device, after receiving an update notification, the update program is downloaded to the free storage area. Then, after the download is completed, the existing program is switched to the update program.

For this reason, conventional software-equipped devices require a separate storage area in accordance with the software update scale, which increases the cost of the storage area.

The present invention has been made to solve the above-described problems. The first program for controlling a device connected to a device, and the size of the program is smaller than that of the first program. A storage unit storing a second program including a part of the plurality of functions of the program, and the first unit when connected to the storage unit and receiving an update request for the first program A control unit that stops the execution of the program, overwrites and updates the first program, and executes the partial function of the second program during the update of the first program. Providing software-equipped equipment with features.

The present invention can update software with a small storage area.

1 is a system configuration diagram according to an embodiment of the present invention. It is a hardware block diagram of the software mounting apparatus in embodiment of this invention. It is a software block diagram which operate | moves on the software mounting apparatus in Embodiment 1 of this invention. It is a block diagram of the program switching part in embodiment of this invention. It is a flowchart which shows the operation | movement of the whole process in Embodiment 1 of this invention. It is the figure which showed the example of extraction of the periodic execution command sequence in Embodiment 1 of this invention. It is an example of the execution instruction in Embodiment 1 of this invention. It is a figure which shows switching from the main program in Embodiment 1 of this invention to a subprogram. It is a software block diagram which operate | moves on the software mounting apparatus in Embodiment 2 of this invention. It is a flowchart which shows the operation | movement of the whole process in Embodiment 2 of this invention. It is the figure which showed the example of extraction of the periodic output pattern in Embodiment 2 of this invention. It is a figure which shows the switching from the main program in Embodiment 2 of this invention to a subprogram. It is a software block diagram which operate | moves on the software loading apparatus in Embodiment 3 of this invention. It is a flowchart which shows the operation | movement of the whole process in Embodiment 3 of this invention. It is the figure which showed the example of extraction of the periodic output pattern containing NOP in Embodiment 3 of this invention. It is the figure which showed the example of the addition of the expansion process with respect to the no-processing time in the periodic execution command sequence in Embodiment 3 of this invention.

Embodiment 1 FIG.
FIG. 1 is a system configuration diagram according to the present embodiment. The system in the present embodiment includes a software-equipped device 1, a software distribution server 2, and a network 3. The software-equipped device 1 is a built-in device such as white goods or a personal computer, and downloads software and updates the downloaded software through a network. The software distribution server 2 has a function of distributing an update program (software) to the software-equipped device 1 through the network. The network 3 is a communication means such as the Internet. The network 3 may be wireless or wired.

FIG. 2 is a hardware configuration diagram of the software-equipped device 1 in the present embodiment. In FIG. 2, the software-equipped device 1 includes a CPU 10 (Central Processor Unit), ROM 11 (Read Only Memory), RAM 12 (Random Access Memory), a communication module 13, a device 14, an LED (Light Emitting Diode) 15, and an electronic buzzer. 16, a timer 17, and a power switch 18.

The ROM 11 provides an area for storing a program instruction sequence and fixed data. It corresponds to a flash memory and the like, and the written program content can be rewritten.
The RAM 12 provides a work area for storing and referring to data used by the program.

The communication module 13 transmits and receives data between communication devices through a network infrastructure such as the Internet.

The device 14 is a device that the software-equipped device 1 has. Examples of devices include stepping motors and PWM (Pulse Width Modulation) motors that are used to operate refrigerator fans and compressors, air conditioner fans and louvers, and the like. Note that although only one device 14 is described in the software-equipped device 1 in FIG. 2, it may be configured by a plurality of devices.

The LED 15 is a semiconductor element that notifies the user of the operation status of the device by light pattern output such as lighting and blinking.

The electronic buzzer 16 is a device that notifies the user of the operation status of the device by ringing an electronic sound.

The timer 17 measures the timing at which a program command is executed, the output timing to the device 14 and the like in addition to the time.

The power switch 18 is a button for operating the power source of the software-equipped device 1, and the switch may be mounted on the main body or a remote controller.

FIG. 3 is an overall configuration diagram of software operating on the software-equipped device 1 in the present embodiment. These software are stored in the ROM 11 or the RAM 12 and executed by the CPU 10 as a control unit. In FIG. 3, a main program (first program) 100 is a program that characterizes the function of the software-equipped device 1 as a product. Usually, the main program 100 is a device connected to the software-equipped device 1. Control is in progress.

The subprogram 200 has an instruction to execute device control instead of the main program 100 when an update request to the main program 100 is generated, and includes a software module for updating the main program 100.

The program switching unit 300 receives an update request for the main program 100 from the outside such as the software distribution server 2 or makes an inquiry about the update request to the outside. As a result, when the update request is received, the subprogram 200 Is a program that instructs updating. In addition, the program switching unit 300 has a function of ending the control of the device executed by the subprogram 200 after the update to the main program 100 is completed and returning the device control to the main program 100 replaced with the update program. have.

First, the details of the main program 100 will be described. The OS 110 is basic software for operating the main program 100. The application operating on the OS 110, the application a (120a), the application b (120b), and the application n (120n) are programs in which the functions of the software-equipped device 1 are installed. A driver a (130a), a driver b (130b), and a driver m (130m) incorporated in the OS 100 are device drivers and programs for controlling the hardware (device) of the software-equipped device 1. is there.

Next, the details of the subprogram 200 will be described. The scheduling unit 210 receives an operation instruction from the program switching unit 300. There are two types of operation instructions. The first is an analysis instruction for examining the operation status of the main program 100. The second is an update execution instruction for executing the control of the device in place of the main program 100 during the update work and the update work in which the content of the main program 100 is replaced with the update program.

When the analysis instruction is received, the scheduling unit 210 is called. The called analysis unit 220 collects execution instructions of the main program 100 and pattern-determines the periodicity of the execution instructions from the collected execution instruction sequence (execution instruction sequence). Based on this pattern determination, the analysis unit 220 extracts a group of instructions that the CPU 10 executes periodically, and stores the instruction group in the storage area 230 as a periodic execution instruction sequence. After the analysis is completed, the analysis unit 220 notifies the program switching unit 300 of the completion.

The storage area 230 is an area (storage section) for storing the periodic execution instruction sequence extracted by the analysis section 220, and is a storage area on the ROM 11 and RAM 12 that is reserved by the subprogram 200.

Next, a module related to the update of the main program 100 will be described. The scheduling unit 210 calls the execution unit 240 and the update unit 250 when receiving an update execution instruction. The execution unit 240 is a module that executes a periodic program (second program) previously extracted from the software operation of the main program 100 and stored in the storage area 230. On the other hand, the update unit 250 receives the update program from the software distribution server 2 via the communication module 13 while the execution unit 240 is executing the periodic program extracted from the software operation of the main program 100. It is a module that performs rewriting and updating.

Next, details of the program switching unit 300 will be described. The program switching unit 300 is a program for instructing the main program 100 and the subprogram 200 to start or stop. The program switching unit 300 is called first when the software-equipped device 1 is activated, and causes the CPU 10 to execute the main program 100.

FIG. 4 is a diagram illustrating a software configuration inside the program switching unit 300. In FIG. 4, an update request receiving unit 310 receives an update request that the software distribution server 2 transmits to the software-equipped device 1 before distributing the update program. This reception is performed via the communication module 13 of FIG. When receiving the update request, the update request receiving unit 310 sends an analysis instruction to the subprogram 200 via the program start / stop unit 320. The program start / stop unit 320 sends an operation instruction to the main program 100 and the subprogram 200.

The subprogram 200 that receives the analysis instruction executes the analysis of the main program 100, and returns an analysis completion notification when the analysis is completed. Receiving this analysis completion notification, the cycle timing synchronization unit 330 sends a stop instruction to the main program 100 and an update execution instruction to the subprogram 200 via the program start / stop unit 320.

Upon receiving the update execution instruction, the subprogram 200 executes the update process of the main program 100 and transmits an update completion notification to the update completion receiving unit 340 when the update is completed. Receiving this notification, the update completion receiving unit 340 sends a stop instruction to the subprogram 200 and sends an execution instruction to the main program 100 via the program start / stop unit 320. Note that since the program switching unit 300 starts the main program 100 immediately after startup, the main program 100 can be started by rebooting the software-equipped device 1.

Next, the operation of the overall processing in the first embodiment will be described with reference to FIG. Here, a case where a software update notification is received from the software distribution server 2 while the software-equipped device 1 is executing device control will be described as an example.

In this embodiment, it is assumed that the software distribution server 2 obtains an update program for the main program 100 from the outside (not shown). The acquisition method may be received from another network, or may be acquired via an external storage device or medium.

Further, the software distribution server 2 manages the version of the main program 100 used in the software-equipped device 1, and when an update program is obtained, the software distribution server 2 performs a comparison and updates only when it is determined that an update is necessary. Assume that a request is sent to the software device 1.

Note that the software distribution server 2 may send all received update programs to the software-equipped device 1 so that the software-equipped device 1 may determine whether update is necessary.
In addition to this, the software distribution server 2 may send an update request only when the software-equipped device 1 inquires about the presence or absence of the update program and the update program exists.

First, the update request receiving unit 310 receives a software update request from the software distribution server 2 (S101). Next, the update request receiving unit 310 sends an analysis instruction for examining the operation status of the main program 100 to the subprogram 200 via the program start / stop unit 320.

As a method of sending an analysis instruction to the subprogram 200, the scheduling unit 210 may be called together with an argument that can be determined as an analysis instruction from the program start / stop unit 320. Also, a memory area that can be accessed in common by the program start / stop unit 320 and the scheduling unit 210 is prepared, the scheduling unit 210 polls this area, and the program start / stop unit 320 changes the value of this area. Analysis instructions may be sent to the subprogram 200.

The scheduling unit 210 that has received the analysis instruction calls the analysis unit 220. The analysis unit 220 analyzes the main program 100. In performing the analysis, the analysis unit 220 records an execution instruction operating in the main program 100 (S102).

In order to obtain information on an execution instruction operating in the main program 100, a method of referring to a memory storing a machine language from a program counter register of the CPU 10 or an application (120a to 120n operating in the main program 100) ) And an execution instruction of the OS 110 to output a machine language to a memory area or a log file and refer to it.

The analysis unit 220 analyzes the recorded sequence of execution instructions (execution instruction sequence), extracts a cyclic execution instruction sequence from the execution instruction sequence, and stores it in the storage area 230 (S103). The analysis of the execution instruction sequence will be described with reference to FIGS.

FIG. 6 is a diagram showing an example of extraction of a cyclic execution instruction sequence in the present embodiment. In the following, a method for detecting periodically executed instructions and extracting instructions to be executed during update processing will be described. However, the present invention is not limited to this, and instructions corresponding to predetermined functions are described below. It is also possible to use a method of detecting an instruction and extracting an instruction. For example, an address indicating a specific range in the main program 100 can be stored in the ROM 11 in advance, or the software-equipped device 1 can obtain from the distribution server 2 to specify and extract a command.

In FIG. 6, execution instructions 610a to 610m are machine languages of the main program 100 executed by the CPU 10. FIG. 6 shows a state in which execution is performed sequentially from the left along the execution time. A group of a plurality of execution instructions is called an execution instruction sequence. In the example of FIG. 6, a group of execution instructions 610a to 610m is called an execution instruction sequence 610. The periodic execution instruction sequence 620 is an execution instruction sequence that is periodically repeated as a result of analyzing the execution instruction sequence 610 (execution instructions 610a to 610m).

As a method of collecting execution instructions from the main program 100, an instruction executed by an application (120a to 120n) operating in the main program 100 may be output to a log, and the analysis unit 220 may refer to it. . In addition, an instruction executed by the OS 110 may be output to a log, and the analysis unit 220 may refer to it.

The analysis unit 220 records and analyzes the collected execution instructions. As shown in FIG. 6, A, B, C, D, A, B, C, D, A, B, C, D, A, etc., the same repetitive pattern continues for a certain period of time. If detected, for example, in this case, A, B, C, and D are extracted as the periodic execution instruction sequence 620. Thereafter, the analysis unit 220 stores the extracted periodic execution instruction sequence 620 in the storage area 230.

The fixed period and the number of repetitive patterns to be analyzed depend on how periodically all the applications (120a to 120n) executed on the software-equipped device 1 operate, so statistically investigated in advance. For example, a value such as a period of 100 milliseconds and the number of patterns of 5 is used as a predetermined condition.

Note that as a pattern extraction algorithm for determining periodicity, an algorithm such as a pattern recognition algorithm used in natural information processing or the like may be used.

Fig. 7 shows an example of execution instructions. The collected data 701 is a collection of execution instructions collected by the analysis unit 220, and the analysis data 702 indicates a periodic execution instruction sequence extracted after the analysis unit 220 analyzes the collected data 701.

The left column of the collected data 701 and the analysis data 702 is an execution ID for identifying an execution command. In the execution instruction column of FIG. 7, for ease of understanding, it is written in assembly language (mnemonic), but actually machine language is substituted.

For example, when the same execution instruction is repeatedly generated continuously as in the collected data 701, the analysis unit 220 extracts the execution ID 0x3502 to execution ID 0x3505 and the execution ID 0x3506 to execution ID 0x350B as a cyclic execution instruction sequence. To do.

When the cyclic execution instruction sequence cannot be extracted (S104: NO), the execution instruction is recorded (S102) again. Although not shown in FIG. 5, when the execution instruction recording (S102) is repeated, the execution instruction may be executed after waiting for a while, not immediately. The time to wait for a while may be determined in advance, or a value sent together with the update program may be used.

When the cyclic execution instruction sequence can be extracted (S103) (S104: YES), the analysis unit 220 sends an analysis completion notification to the program switching unit 300. Receiving the analysis completion notification, the cycle timing synchronization unit 330 switches from the main program 100 to the subprogram 200. First, the cycle timing synchronization unit 330 calculates the timing for switching the execution position from the main program 100 to the cycle execution command sequence based on the information of the cycle execution command sequence received from the analysis unit 220 (S105).

There are the following methods for calculating the timing. As an analysis completion notification or in addition to an analysis completion notification, the analysis unit 220 executes an execution ID of the first execution instruction in the cyclic execution instruction sequence, a time when the first execution instruction is executed in the main program 100, and The period time required for the execution of the cyclic execution instruction sequence is sent. Note that when this method is used, when the analysis unit 220 collects execution instructions of the main program 100, it is also necessary to record the time.

The cycle timing synchronization unit 330 calculates the next time from the current time (eg, t1), the time (t0) last executed by the main program 100, and the cycle time (tc) required to execute the cycle execution instruction sequence. The predicted time (tn) of the first execution instruction of the cyclic execution instruction sequence to be executed is calculated. Then, the cycle timing synchronization unit 330 performs switching from the main program 100 to the subprogram 200 at the predicted time (tn). (S106).

FIG. 8 shows an example of switching from the main program 100 to the subprogram 200. In the example of FIG. 8, the execution of the main program 100 is stopped at the time tn obtained by adding the period time tc to the time t0 last executed by the main program 100, and the subprogram 200 instead. The execution instruction of the periodic execution instruction sequence 620 is issued.

There is a method of using the program start / stop unit 320 when the cycle timing synchronization unit 330 switches from the main program 100 to the subprogram 200. The program start / stop unit 320 sends a stop operation instruction to the main program 100.

As a method of sending a stop operation instruction, an application for receiving a stop instruction may be prepared in advance on the main program 100 side and called from the program start / stop section 320, or the program switching section 300 and the main program may be called. A memory area that can be accessed in common with 100 is prepared, and the program start / stop unit 320 may send a stop operation instruction to the main program 100 by changing the value of this area.

Then, for the subprogram 200, the program start / stop unit 320 sends an update execution instruction to the scheduling unit 210.

The scheduling unit 210 that has received the update execution instruction executes the execution unit 240 and the update unit 250 at the same time. One execution unit 240 refers to the storage area 230 and starts executing the cyclic execution instruction sequence. Note that the execution unit 240 continues to execute the cyclic execution instruction sequence repeatedly.

The other update unit 250 downloads the update program via the communication module 13 and updates the main program 100 by overwriting (S107). While performing the update, the execution unit 240 executes a part of the processing required by the software-equipped device 1, so that a new memory is not required for the update program, and the update program is directly added to the main program 100. It can be downloaded.

After the download of the update program is completed, the update unit 250 sends a completion notification to the program switching unit 300 (S108).

Upon receipt of the completion notification, the program switching unit 300 stops the device (S109) and restarts the software-equipped device 1 (S110). Note that, depending on the device 14 connected to the software-equipped device 1, for example, in the case of a device that requires time for activation, the operation of the device 14 is temporarily interrupted, so the program switching unit 300 automatically restarts. Instead, the restart timing may be adjusted by manually operating the power switch 18.

After the software-equipped device 1 is restarted, the program switching unit 300 starts the updated main program 100 (S111).

By adopting the configuration as described above, a part of the program is copied to the storage device and used by focusing on some of the functions of the main program 100 that need to be executed even during the update. Therefore, the copied program is smaller in size than the main program 100, and the capacity of the storage area necessary for software update can be saved. In addition, the update operation can be performed while continuing the operation on the device 14.

In this embodiment, after receiving the software update request from the software distribution server 2, the analysis unit 220 extracts the cyclic execution instruction sequence. However, the cyclic execution instruction sequence is extracted before the update request is received. You can go there. Further, while the analysis unit 220 does not extract the periodic execution instruction sequence and updates the main program 100, the cyclic execution instruction sequence necessary for the software-equipped device 1 to control the device 14 is stored in the storage area 230 in advance. It may be stored. At this time, the timing at which the cycle timing synchronization means 330 sends a stop instruction to the main program 100 is also stored in the storage area 230.
The periodic execution instruction sequence necessary for the software-equipped device 1 to control the device 14 may be stored in the storage area 230 at the time of factory shipment, or may be written later. Alternatively, the update unit 250 may receive the update program received from the software distribution server 2 and store it in the storage area 230.

Embodiment 2. FIG.
In the present embodiment, a case will be described in which the operation of the device 14 connected to the device is referred to in order to extract a process that repeatedly operates.

FIG. 9 is an overall configuration diagram of software that operates on the software-equipped device 1 in the present embodiment.

In FIG. 9, a periodic output analysis unit 920 is software for monitoring the operation of the device driver of the main program 100, determines its periodicity from its execution order and timing, extracts a periodic output pattern, and executes the same periodic output A possible periodic execution instruction sequence is stored in the storage area 230.

9, the same reference numerals as those in FIG. 3 represent the same or corresponding parts, and operations other than the periodic output analysis unit 920 perform the same operations as those described in FIG. 3 used in the first embodiment.

Next, the overall operation in the second embodiment will be described with reference to FIG. As in the case of the first embodiment, a case where a software update notification is received from the software distribution server 2 while the software-equipped device 1 is executing will be described as an example.

The operation in the second embodiment is the same as the procedure described in the first embodiment, and only the difference from the operation described in the first embodiment will be described below.

The update request receiving unit 310 sends an analysis instruction for examining the operation status of the main program 100. This instruction can be realized by sending an analysis instruction by calling the scheduling unit 210 from the program start / stop unit 320, or can be commonly accessed by the program start / stop unit 320 and the scheduling unit 210. A memory area is prepared, the scheduling unit 210 polls this area, and the program start / stop unit 320 changes the value of this area, thereby sending an analysis instruction to the subprogram 200. You can also.

Upon receiving the output analysis instruction, the scheduling unit 210 calls and executes the periodic output analysis unit 920. The periodic output analysis unit 920 analyzes the main program 100. In performing the analysis, the periodic output analysis unit 920 monitors the device drivers (driver a to driver m) operating in the main program 100 and records their execution order (output instructions) and their execution times ( S202).

To obtain output instruction information, you can poll the memory area to which the device register for controlling the device is allocated and refer to it, or generate an interrupt when controlling the device. There is a method of designing a driver and receiving it by the periodic output analysis unit 920.

The recording destination of the periodic output may be the memory area on the RAM 12 temporarily prepared by the periodic output analysis unit 920 or the storage area 230.

The periodic output analysis unit 920 analyzes the recorded periodic output, extracts an execution instruction sequence that can execute the same periodic output from a list of execution instruction sequences prepared in advance, and stores it as a periodic execution instruction sequence Store in the area 230 (S203). The cyclic execution instruction sequence is the same as in the first embodiment.
Note that only one execution instruction sequence prepared in advance may be used as a cyclic execution instruction sequence that is repeatedly executed. When repeatedly executing, the time, the address of the memory to be written, the value to be written, and the time to write the value may be passed as arguments to the execution instruction sequence. The analysis of periodic output will be described with reference to FIG.

FIG. 11 is a diagram showing an example of extraction of a periodic output pattern in the present embodiment. In FIG. 11, device output records 1110a to 1110i are records of output patterns for the device 14 and their execution times. FIG. 11 shows a state in which execution is performed sequentially from the left along the execution time. The periodic output pattern 1120 is an output pattern that is periodically repeated as a result of analyzing the device output records 1110a to 1110i. The execution instruction sequence list 1130 is a list of execution instruction sequences for executing a specific output pattern. Advance what the output instruction previously examined whether performed, are prepared an execution command sequence to perform the same output. In FIG. 11, examples of the execution instruction sequence f_E for the output pattern α, the execution instruction sequence f_F for the output pattern β, the execution instruction sequence f_G for the output pattern γ, and the execution instruction sequence f_H for the output pattern Δ are shown. Show.

For example, if there are a motor X and a motor Y as the device 14, the state of outputting 1 to the motor X is a millisecond (this is an output pattern α), and the state of outputting 2 to the motor X. b milliseconds (this is assumed to be output pattern β), the state that 3 is output to motor Y is c milliseconds (this is assumed to be output pattern γ), output pattern α, output pattern β, output pattern γ, output When the same repeated output pattern is detected continuously for a certain period of time, such as pattern α, output pattern β, and output pattern γ (example of device output recording in FIG. 11). Output pattern α and output pattern β The output pattern γ is extracted as the periodic output pattern 1120.

The periodic output analysis unit 920 creates a periodic execution instruction sequence combination 1140 for executing the extracted periodic output pattern 1120 by referring to the execution instruction sequence list 1130. Thereafter, the cycle output analysis unit 920 stores the combination 1140 of cycle execution instructions in the storage area 230.

The fixed period and the number of repetitive output patterns to be analyzed depend on how periodically all the device drivers executed in the software-equipped device 1 operate. For example, it is used as a condition where a value such as a period of 100 milliseconds and an output pattern number of 5 is given.

Also, the access time to the device may vary depending on the device status (environment such as temperature and humidity, and aging degradation). Therefore, for example, in the above example, if the output patterns of α, β, and γ are observed, even if there is a slight increase / decrease in each execution period, it is allowed. Since the allowable range depends on the device connected to the software-equipped device 1, it is used as a condition in which a value such as an increase or decrease of 1% is given as a result of statistical investigation in advance.

Returning to the description of the processing flow of FIG. 10, when the periodic output pattern cannot be specified (S204: NO), the device drivers (driver a to driver m) operating in the main program 100 are monitored and their execution order is monitored. (Output command) and its execution time are recorded again (S202). Although not shown in FIG. 10, when the execution order and the recording of the execution time (S202) are repeated, they may be executed after waiting for a while rather than immediately. The time to wait for a while may be determined in advance, or a value sent together with the update program may be used.

When the periodic output pattern is identified (S203) (S204: YES), the periodic output analysis unit 920 sends an analysis completion notification to the program switching unit 300. Receiving the analysis completion notification, the cycle timing synchronization unit 330 switches from the main program 100 to the subprogram 200. First, the cycle timing synchronization unit 330 calculates timing for switching execution from the main program 100 to the execution unit 240 based on the information of the cycle output pattern received from the cycle output analysis unit 920 (S205).
There are the following methods for calculating the timing. The periodic output analysis unit 920 includes, as an analysis completion notification or in addition to an analysis completion notification, an execution ID of the first execution instruction sequence of the cycle execution instruction sequence combination 1140 and a device corresponding to the first execution instruction sequence The time when the output 14 is executed by the main program 100 and the period time required for the execution of the cycle execution instruction sequence are sent. Note that when this method is used, when the output pattern of the main program 100 is collected by the periodic output analysis unit 920, the time must also be recorded.

The cycle timing synchronization unit 330 is configured to execute the current time (for example, t1), the time (t0) when the output pattern was last output in the main program 100, and the cycle time required to execute the combination 1140 of the cycle execution instruction sequence ( From tc), the predicted time (tn) of the first execution instruction of the cyclic execution instruction sequence to be executed next is calculated. Then, the cycle timing synchronization unit 330 performs switching from the main program 100 to the subprogram 200 at the predicted time (tn). (S206).

FIG. 12 shows an example of switching from the main program 100 to the subprogram 200. In the example of FIG. 12, the execution of the main program 100 is stopped at the time tn obtained by adding the period time tc to the time t0 last executed in the main program 100, and α is output to the subprogram 200 instead. An instruction to execute a combination of periodic execution instruction sequences 1140 is issued.

Upon receiving this instruction, the scheduling unit 210 executes the execution unit 240 and the update unit 250 at the same time. The execution unit 240 refers to the storage area 230 and starts executing the cyclic execution instruction sequence. Note that the execution unit 240 continues to execute the cyclic execution instruction sequence repeatedly.

The subsequent processing is the same as the procedure described in the first embodiment.

にする With the above configuration, it is possible to save the capacity of the storage area required for software update. In addition, the update operation can be performed while continuing the operation on the device.

In the present embodiment, the periodic output analysis unit 920 extracts the periodic output pattern after receiving the software update request from the software distribution server 2, but the periodic output pattern is extracted before the update request is received. You can go there.

Embodiment 3 FIG.
In the present embodiment, a case will be described in which a no-processing period is analyzed with respect to the periodic execution instruction sequence 620 analyzed by the analysis unit 220 and an extended processing program statically prepared by a user is inserted in advance.
By inserting an extended processing program prepared statically by the user, it is possible to execute the extended processing program while the periodic execution instruction sequence 620 is being executed.
For example, if an external interrupt reception such as a remote control or an LED flashing program is statically prepared as an extended processing program, the remote control reception processing and LED flashing processing may be performed while the periodic execution instruction sequence 620 is being executed. it can.

FIG. 13 is an overall configuration diagram of software operating on the software-equipped device 1 in the present embodiment.

In FIG. 13, an extended processing storage area 2040 is an extended processing program that is statically prepared in advance by the user. The extended processing program is a program that is desired to be further executed during execution of the periodic execution instruction sequence 620, and may be a program for accepting external interrupts such as the above-mentioned remote control or a blinking program for LEDs. The extended processing program prepared in the extended processing storage area 2040 may be obtained via an external storage device or a communication module.

In FIG. 13, the no-process period extraction unit 2010 extracts a period (no-process period) in which no-process continues continuously from the periodic execution instruction sequence stored in the storage area 230 and stores it in the no-process-period storage area 2020. It is.
In this embodiment, the non-processing period storage area 2020 and the extended processing storage area 2040 are configured differently from the storage area 230. However, the contents stored in the non-processing period storage area 2020 and the extended processing storage area 2040 May be stored in the storage area 230.

In FIG. 13, the extended process division insertion unit 2030 refers to the no-process period from the no-process period storage area 2020, divides the extended process instruction sequence stored in the extended process storage area 2040 by the no-process period length, and Duplicate the cyclic execution instruction sequence by the number, and for each duplicated cyclic execution instruction sequence, duplicate the divided extended processing instruction sequence in its non-processing period part, and copy those extended processing instruction sequences to one cyclic execution instruction It is stored in the storage area 230 as a column.

In FIG. 13, the same reference numerals as those in FIG. 3 represent the same or corresponding parts, except for the non-processing period extraction unit 2010, the non-processing period storage area 2020, the extended processing division insertion section 2030, and the extended processing storage area 2040. Performs the same operation as that described in FIG. 3 used in the first embodiment.

Next, the overall operation in the third embodiment will be described with reference to FIG. As in the case of the first embodiment, a case where a software update notification is received from the software distribution server 2 while the software-equipped device 1 is executing will be described as an example.

The operation in the third embodiment is the same as the procedure described in the first embodiment, and only the difference from the operation described in the first embodiment will be described below.

The non-processing period extraction unit 2010 extracts a non-processing period from the cyclic execution instruction sequence stored in the storage area 230.
The no-processing period refers to a period in which a situation in which the CPU is not performing calculation processing or accessing a register, such as a NOP (No Operation) instruction in the assembler.
The non-processing period extraction unit 2010 stores this non-processing period information in the non-processing period storage area 2020 (S301). With the above processing, it can be understood at which timing no processing is performed in the cyclic execution instruction sequence.
The non-processing period extraction unit 2010 calls the extended processing division insertion unit 2030 after storing the non-processing period information in the non-processing period storage area 2020.

The extended process dividing / inserting unit 2030 copies the extended process program stored in the extended process storage area 2030 to the non-process period of the cyclic execution instruction sequence and stores it in the storage area 230 as a new cyclic execution instruction sequence. Yes (S302).

After completing the download of the update program, the update unit 250 checks the cyclic execution instruction sequence stored in the storage area 230 and determines whether there is an external interrupt process in the interrupt process (S303). The determination may be performed by directly checking the inside of the cyclic execution instruction sequence, or when the extended processing division insertion unit 2030 updates the cyclic execution instruction sequence, information indicating that fact is included in a part of the storage area 230. It may be stored.
For example, when it is desired to interrupt and restart a periodic execution by an external interrupt from a remote controller or the like, a process for receiving an external interrupt signal is defined as an extended process.
When the update unit 250 receives an external interrupt signal (S304: YES), the update unit 250 sends a download completion notification of the update program to the program switching unit 300, and the program switching unit 300 stops the device and restarts it.
When the external interrupt signal is not received (S304: NO), the update unit 250 waits until an external interrupt signal is received. Although not shown, the update unit 250 continues to wait until an external interrupt signal is received, so that the process automatically proceeds to S108 when a preset time has elapsed. Also good.

FIGS. 15 and 16 are diagrams illustrating an example of extraction of a non-processing period in the non-processing period extraction unit 2010 of this embodiment and addition of an extension process to the non-processing period.
In FIG. 15, execution instruction sequences 2110a to 2110m are execution instruction sequences obtained by collecting execution instructions of the main program 100, and NOPs appear periodically such as 2110c, 2110g, and 2110k.
The method for extracting the repetitive pattern is the same as in the first embodiment, and the extracted periodic execution instruction sequence is defined as a first periodic execution instruction sequence 2120. The first periodic execution instruction sequence 2120 includes NOP2120 nop.

The details when the no-process period extraction unit 2010 extracts the no-process period and stores the extension process will be described with reference to FIG. In the present embodiment, as an example of the extension process stored in the extension process storage area 2040, there is an extension process 2130 having an instruction length of topt.
The no-process period extraction unit 2010 extracts the continuous NOP process continuous period 2120 nop from the extracted periodic execution instruction sequence 2120 and stores it in the no-process period storage area 2020.

Next, the extended process division inserting unit 2030 refers to the extended process 2130 stored in the extended process storage area 2040 and compares the no-process period 2120 nop with its instruction length.
At this time, when the instruction length topt of the extension process 2130 is longer than the instruction length tnop of the no-process period 2020 nop, the extension process 2130 is time-divided by tnop.
For example, it is assumed that the time-divided expansion processes are stored as the expansion process 2131 (expansion 1) and the expansion process 2132 (expansion 2), respectively.

Next, the extended process dividing / inserting unit 2030 duplicates the cyclic execution instruction sequence 2120 and prepares it for the number of time divisions (two in this case), and creates the cyclic execution instruction sequence 2141 and the cyclic execution instruction sequence 2142.
Then, the extension process 2131 and the extension process 2132 are duplicated for the no-process period 2120 nop of each cyclically executed instruction sequence. The non-processing period 2120 nop of the cyclic execution instruction sequence 2141 and the extended processing 2131 are exactly the same because the processing time is the same, but the non-processing period 2120 nop of the cyclic execution instruction sequence 2141 and the extended processing 2132 do not match the processing time. In the illustrated example, NOP2133 remains.

Next, the extended process dividing / inserting unit 2030 concatenates the copied periodic execution instruction sequences and stores them in the storage area 230. In the example of the present embodiment, the cyclic execution instruction sequence 2141 and the cyclic execution instruction sequence 2142 are concatenated to generate one cyclic execution instruction sequence 2150 and save it in the storage area 230.

It should be noted that an algorithm such as a pattern recognition algorithm used in natural information processing or the like may be used as a pattern extraction algorithm for extracting a non-processing period.

Further, an arbitrary program of the user may be stored in the expansion process. For example, in the case of a program that causes the LED to blink, it is possible to cause the LED that accompanies the product to blink while the update program is being downloaded.
For example, if the program receives a remote control interrupt and restarts (jumps to the reset vector), after downloading the update program, it receives the remote control signal, stops periodic execution at the user's timing, and restarts Is possible.

The extension process may be an arbitrary program of the user, but it must not be a program dependent on the OS or existing library, such as an OS system call or an existing library call. This is because the periodic processing is composed only of a periodic execution instruction sequence and does not include processing such as an OS.

With the configuration as described above, it is possible to save the capacity of a storage area necessary for software update. In addition, the update operation can be performed while continuing the operation on the device. It is also possible to execute an arbitrary program prepared in advance during software update.
In the above description, the storage area 230 is the storage unit. However, the areas for storing the main program 100, the subprogram 200, and the program switching unit 300 may be combined as a storage unit.

1 software-equipped device, 2 software distribution server, 3 network, 10 CPU, 11 ROM, 12 RAM, 13 communication module, 14 device, 15 LED, 16 electronic buzzer, 17 timer, 18 power switch, 100 main program, 110 OS, 120a-120n application, 130a-130m device driver, 200 subprogram, 210 scheduling unit, 220 analysis unit, 230 storage area, 240 execution unit, 250 update unit, 300 program switching unit, 310 update request receiving unit, 320 program start / stop Unit, 330 periodic timing synchronization unit, 340 update completion receiving unit, 610a to 610m execution instruction sequence, 620 periodic execution instruction sequence, 701 analysis unit 22 , 702, table showing periodic execution instruction sequences extracted after analysis by the analysis unit 220, 920 periodic output analysis unit, 1110a to 1110i device output record, 1120 periodic output pattern, 1130 List of execution instruction sequences, 1140 Combination of cyclic execution instruction sequences, 2010 No processing period extraction unit, 2020 No processing period storage area, 2030 Extended processing division insertion section, 2040 Extended processing storage area, 2110a to 2110h Execution instruction sequence, 2120 nop No processing Period, 2130 Extended processing instruction sequence, 2131 Extended processing instruction sequence, 2132 Extended processing instruction sequence, 2133 NOP, 2141 Periodic execution instruction sequence, 2142 Periodic execution instruction sequence, 2150 Periodic execution instruction sequence

Claims

A first program for controlling a device connected to the device, and a second program having a smaller program size than the first program and including a part of the plurality of functions of the first program A storage unit storing
When receiving an update request for the first program connected to the storage unit, the execution of the first program is stopped, the first program is overwritten and updated, and the first program is being updated. And a control unit that executes the partial function of the second program.
The control unit refers to an instruction executed by the first program to control a device, extracts a repeatedly executed instruction as a periodic execution instruction sequence, and extracts the extracted periodic execution instruction sequence as the first execution sequence. The software-equipped device according to claim 1, wherein the software is stored in the storage unit as a second program.
The control unit detects an operation that is repeatedly performed on the device, extracts a plurality of instructions corresponding to a process that repeatedly operates as a periodic execution instruction sequence, and extracts the extracted periodic execution instruction sequence as the second execution sequence. The software-equipped device according to claim 1, wherein the software-installed device is stored in the storage unit as a program.
The control unit extracts a no-process period during which no processing is performed for the periodic execution instruction sequence, and inserts an extension process to be executed during an update into the no-process period. 2. Software-equipped device according to 2.
5. The software-equipped device according to claim 4, wherein the control unit divides the extended processing and inserts the extended processing into a plurality of the non-processing periods.
Transmitting an update request for a first program that controls a device connected to the device;
While the first program is being updated, extracting a periodic execution instruction sequence for continuing only a part of the control of the device and storing it in a storage unit;
When the device receives the update request, the device stops executing the first program and executes the periodic execution instruction sequence;
Transmitting an update program of the first program to the device executing the periodic execution instruction sequence;
A software updating method comprising: after the download of the update program, the device stops executing the cyclic execution instruction sequence and executes the first program replaced by the update program. .