JPH0962499A - Program file update device - Google Patents

Abstract

(57) [Summary] [Purpose] The control microprogram is replaced while the online operation continues, and the file update judgment function and the old version promptly occur when a failure occurs to prevent the occurrence of a failure caused by the update of the control microprogram. To have a function to return to. The processor 20 is closed while the processors 21 to 23 continue to operate, the control microprogram stored in the new program area 31 of the auxiliary storage device 30 is loaded into the processor 20, and the processor 20 is recovered. The same processing is repeated for the processors 21 to 25, and the program update can be completed in all the processors while continuing the online operation. When incorporating a new microprogram from the floppy disk drive 33, the processor 20-
The version determination and program consistency are executed with the control micro program running at 25.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a program file update in an information processing apparatus which has a function of maintaining and managing the program in a control unit composed of a plurality of processors and memories which operate by the program and which can update the program file. Regarding the device.

[0002]

2. Description of the Related Art Conventionally, as disclosed in Japanese Patent Laid-Open No. 6-95988, a memory area for storing a program executing information communication processing internally without disconnecting communication processing from the outside is specified. It stopped at the point of interruption and switched to the memory area where the new program was loaded.

According to the present invention, one of a plurality of operating processors is blocked so that another processor can perform I / O with the outside.
By updating the program of the blocked processor during execution of the O process, the new program can be updated without stopping at a specific interruption point.

[0004]

In the conventional method, the host C is used.
In order to update the control program while continuing the processing with the PU, if a failure occurs during the control program update, the processing is stopped at a specific interruption point and necessary from the memory area where the old file is loaded. I had to transfer the transfer information to a specific area and update the file. Therefore, if a failure occurs during file updating, the processing with the outside cannot be continued. In addition, an extra memory resource is required to store the file to be updated.

According to the present invention, since the program is updated by at least one processor while the processing with the host CPU and the external magnetic disk device is being executed, when a failure occurs during the program update, the time when the failure occurs It is possible to return to the state before the update processing that was performed in step 1, and return to the program before the update while continuing the processing with the outside by another processor.

Further, in order to prevent an unexpected program malfunction after updating the program, the compatibility is judged by the label comparison between the version of the operating microprogram and the version of the new microprogram to be updated, and the update is performed. Since this is done, it is possible to prevent malfunctions due to updating of the new program file.

[0007]

In order to achieve the above object,
The following method was adopted.

According to claim 1 of the claims, when a program file is updated, only one processor among a plurality of processors having the same path should be closed and updated on each shared memory. After the program is loaded into the one processor, the blocked processor is restarted to operate with the new program. The same process is repeated by the number of processors to update all the processors to the new program.

According to claim 2 of the claims, when a program file is updated, the program update process is recorded, and during the update work until the update of the program file to all the processors is completed. When a failure occurs due to the operation of the update program for one processor, if the update process is executed as it is, the malfunction of the program may be incorporated into the control device. A return process to the program is prepared, and the original old program stored in another area is reloaded to the processor.

According to claim 3 of the claims, when a program file is updated, the update history of the program is recorded as bitmap data, and compatibility with the file or program to be newly updated, Check the consistency and update only the program that matches the version to be updated.

[0011]

According to the present invention, when updating a program file incorporated in a device, one of a plurality of processors having the same path is blocked, a program to be updated is loaded from a common memory area, and the program is restarted. To update the program. Since only the processor that is updating the program is blocked and the other processors can execute the processing by the old program, the program file can be updated without stopping the processing execution with the host CPU and the external magnetic disk device.

According to the present invention, since the information of the program update processing is recorded, during the update work until the update of the program file to all the processors is completed, it is caused by the operation by the update program to one processor. When a failure occurs, by executing the return processing prepared according to the contents of the update processing that was performed when the failure occurred, the processor processing is not stopped and the old processor Can be restored to the program.

According to the present invention, the condition of whether or not the old and new programs can be updated is determined, and the update process is performed only when the update conditions are met. Therefore, it is possible to prevent defective program operation after the update.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the configuration of a magnetic storage device control system (hereinafter referred to as DKC) which is an embodiment of the present invention. DKC is a channel control microprocessor 20
23 and magnetic disk control microprocessors 24, 2
5, a service processor 10 (hereinafter referred to as SVP) common to the control microprocessors is mounted.

The control microprogram of the DKC is the HDD
It is stored in the new program storage area 31 in 30 and SVP-
It can be accessed from the SVP 10 via the HD interface 34. The SVP 10 can load the control micro program into the control microprocessor via the SVP-processor connection interface 46.

The control microprogram is updated by updating the control microprogram stored in the floppy disk with S.
The VP 10 reads from the floppy disk drive 33 and writes it in the new program storage area 31 in the HDD 30. At this time, the version of the microprogram and the file update (FC) history recorded in the bitmap format are compared to determine whether the update is possible. That is, the program version is given a unique label so that it is possible to know whether or not the file can be updated by this label, and the bit created for each program in the file. You can see the quality of the programs by looking at the history of the map format. If it is determined that the program can be updated, the operating microprogram before updating, which has been in the new program storage area 31 until now, is moved to the old program storage area 32.

The SVP 10 instructs the processor 1 to close the processor, the processor 1 closes the processor-channel connection interface 40, and loads the control microprogram from the new program storage area 31. When the loading of the control microprogram is completed, the SVP 10 instructs the processor 1 to restore the processor, and the processor 1 restores the processor-channel connection interface 40 and restarts the processing with the new control microprogram. At this time, the other processor is executing the process using the old program. In this way, when a failure occurs due to the operation by the update program for one processor during the update work until the update of the program file for all the processors is completed, if the update process is executed as it is, the operation of the program is executed. Since a defect may be incorporated in the control device, a return process to the original program is prepared for each update process, and the original old program stored in the separate area 32 is reloaded to one processor.

The same processing is repeated for the number of microprocessors, and when the processor-magnetic disk connection interface 45 of the processor n25 is recovered and the processing is restarted by the new control microprogram, all the microprocessors of the DKC become the new control microprogram. Will be updated.

If an unexpected failure occurs while the DKC is running in the new control microprogram after the update, the SV
SV via P-processor connection interface 46
P10 detects.

The SVP 10 determines whether or not the failure is due to a new microprogram, and if the new control microprogram is the cause of the failure, the control microprogram stored in the old program storage area 32 before update is stored. Move the program to the new program area. SV
P10 repeats the same processing as that at the time of updating to the processor and returns to the control microprogram before updating.

FIG. 2 is a diagram showing details of version and bitmap determination. Compare the label of the version of the running microprogram with the version of the new microprogram being updated before the microprogram update, and if the new microprogram is not compatible with the running program, the processor Clears the bitmap data that has been recorded in the above and represents the program file update history up to that point with a completely new bitmap.

If it is determined that the new microprogram version is compatible with the running microprogram, a prerequisite bitmap (or a bitmap representing the program file update that must be performed before the new microprogram update). Presence / absence map of the program which is a prerequisite for operating the program) and the program file update history bitmap are compared, and when all the bits in the prerequisite bitmap are in the FC history bitmap, the microprogram is updated. It will be possible.

Since the FC history is also updated when the microprogram is updated, a new bit representing the latest FC history is merged by merging the exchange content bitmap showing only the changed content and the file update (FC) history bitmap. Create a map and overwrite it in the FC history bitmap area of the processor.

[0024]

By shutting down only one of the plurality of processors and updating the control microprogram, the online operation is performed with the minimum resources without interrupting the processing from the host CPU and the external magnetic disk device. The inside micro program can be exchanged.

When it becomes necessary to restore the updated microprogram to the pre-updated microprogram, since the old and new generation 2 microprograms are managed, it is possible to quickly restore the old version even during online operation.

Since the version information is judged before the updating process, the maintenance personnel do not have to compare the versions. Elimination of illegal versions due to human mistakes.

[Brief description of drawings]

FIG. 1 shows a processor and a service processor (SVP)
1 is a configuration diagram of a magnetic storage device control system in which is mounted.

FIG. 2 is a schematic diagram illustrating data used for version determination.

[Explanation of symbols]

10: Service Processor (SVP) 20-25: Processor 30: Auxiliary Storage Device 31: New Program Area 32: Old Program Area 33: Floppy Disk Drive 40-43: Processor-Channel Connection Interface 44-45: Processor-Magnetic Disk Connection Interface 46: SVP-processor connection interface

Claims (3)

[Claims] 1. A processor, an auxiliary storage device, and one or more channel ports for communicating with the host CPU, the processor controlling the channel port according to a program loaded from the auxiliary storage device, and the host CPU. In the magnetic storage device control system that performs the communication process of (1) and the signal transfer process with the external magnetic disk device, one of the processors having the same path is blocked and the remaining processors are operated,
By updating the program file of the closed processor and further updating the program files of other processors in sequence, the existing constituent resources can be used without interrupting the processing with the host CPU and the external magnetic disk device. , A program file updating device in which all processors are updated with a new program file. 2. The method according to claim 1, wherein a failure occurs due to the operation of the update program to one processor during the update work until the update of the program files to all the processors in the plurality of processors having the same path is completed. In this case, the program file updating device is characterized in that the updated one processor is restored to the program before the update without interrupting the processes from the host CPU and the external magnetic disk device. 3. The method according to claim 1, wherein, when updating the program file, the compatibility of the file and the consistency of the program in the file are determined between the file in operation and the file to be updated. Characteristic program file update device.