JP2708471B2 - Processing device control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a CCU (Communicati
The present invention relates to a control method of a processing device such as a CMP in a communication control processing device including an on-control unit (communication control device) and a CMP (CCP Management Processor) having a CPU and a main memory.

[Conventional technology]

When the communication control processor (CCP) is powered on,
Alternatively, when an IMPL (Initial Micro-Program Load) request occurs again, the initial diagnostic program (I-DIAG) and the loader stored in the ROM are all stored in the main memory (MM), and after the storing, the control is performed. The initial diagnosis program is executed from a predetermined address according to a circuit instruction.

The initial diagnosis program performs an instruction test of the CPU, a main memory test, a read / write test of an external storage device, a test of a keyboard / display, etc., confirms the normality of the CMP itself, and then branches to a loader. The loader stores the operating system (OS) in the external storage device in the main memory, and branches to an initial setting unit of the OS. Thereafter, the CMP communicates with the CCU under the control of the OS, and starts normal work.

Conventionally, in such a system, if any CMP failure occurs during the execution of the initial diagnosis program or the loader, the conversation processing cannot be performed because the CMP is not operating under the OS, and the collected There was no way to present the data to the operator or maintenance personnel (CE).

[Problems to be solved by the invention]

Conventionally, the cause of a failure during the execution of the initial diagnostic program and loader is determined by examining the operation contents of the CPU and main memory.
It was very difficult because there was no means for notifying the operator by conversation.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above problems and to provide means for investigating a processing device failure such as a CMP failure before the OS starts up.

[Means for solving the problem]

 FIG. 1 is a diagram illustrating the principle of the present invention.

In FIG. 1, 1 is a communication control processing unit (CCP), 2 is a CCU that controls line data, and 3 is a CCP management processing function.
CMP, 4 is a main memory for storing instructions and data, 5 is a CPU for fetching and executing instructions, 6 is a keyboard / display device used for conversation processing with an operator, and 7 is a ROM.
, A magnetic disk drive, 9 a control circuit for controlling IMPL, interrupts, and the like, 10
Is an initial diagnosis program for executing an initial diagnosis, 11 is a loader for loading an OS on the main memory 4, 12 is a mini OS having a simple conversation processing function with an operator, 13 is an operating system (OS), and 14 is an application. A program 15 is data such as various control tables, 16 is a changeover switch for switching a mode, 17 is an interrupt switch for instructing an interrupt, and 18 is a mode register for storing whether the mode is a user mode or a maintenance mode.

In the present invention, a mini OS 12 is stored in the nonvolatile memory 7 in addition to the initial diagnosis program 10 and the loader 11 for checking the normality of the CMP3.

At the time of the IMPL of the CMP3, the contents of the nonvolatile memory 7 are transferred to the main memory 4 by the control circuit 9, and the CPU 5 is instructed to start the execution from the initial diagnosis program 10.

As the execution mode, there are a user mode in which a normal operation is performed and a maintenance mode in which a debugging function can be used. The mode can be selected by a changeover switch 16. The mode information is set in the mode register 18.

If an error occurs during the execution of the initial diagnosis program 10 or the loader 11, the retry is performed in the user mode, the retry is not performed in the maintenance mode, the control is passed to the mini OS 12, and the keyboard / display device 6 is controlled by the mini OS 12. , The conversation process with the operator is performed.
Also, regardless of the user mode or the maintenance mode, the mini OS 12 is configured to be activated via the control circuit 9 by operating the interrupt switch 17, and also in this case, the maintenance conversation process is performed. ing.

[Action]

According to the present invention, a mini OS 12 including a small-scale control program having a simple conversation processing function is stored in the nonvolatile memory 7 together with the initial diagnosis program 10 and the loader 11, and the execution of the OS 13 is started. In the maintenance mode, the mini OS 12 is started in the maintenance mode depending on the state of the changeover switch 16 when a failure occurs in the CMP3 until the failure occurs in the CMP3. In the mode, the retry is continuously executed as in the conventional case.

Also, regardless of the user mode or maintenance mode, the mini OS 12 is started by the operation of the interrupt switch 17, so if an error such as a permanent loop that CMP3 itself cannot detect occurs, any timing can be set. By operating the interrupt switch 17, the mini OS 12 can carry out the investigation. The activation of the mini OS 12 by the interrupt switch 17 is performed by the initial diagnosis program 10 or the loader
It can be used during the initial debugging of 11 and is also effective for detecting software failures.

〔Example〕

FIG. 2 is an explanatory diagram of a mode judgment in a main memory test according to an embodiment of the present invention, FIG. 3 is an explanatory diagram of a mode judgment in a loader of an embodiment of the present invention, and FIG. Example of control, FIG. 5 is an example of processing of a mini OS according to an embodiment of the present invention, FIG. 6 is an example of display processing according to an embodiment of the present invention, and FIG. The example of a process is shown.

As in the prior art, the initial diagnosis program 10 shown in FIG. 1 starts the system by executing an instruction test of the CPU 5, a test of the main memory 4, a read / write test of the magnetic disk device 8, a test of the keyboard / display device 6, and the like. It has a processing function to check the normality of CMP3 when it is raised. In the present invention, processing for determining the mode instructed by the changeover switch 16 is further added.

Hereinafter, an embodiment of the present invention for mode determination in the main memory test will be described with reference to FIG. The same applies to other tests.

A predetermined start address is set in an address register.

The start address is used as data, and WR is stored in main memory.
ITE.

Take the complement of the start address and set it in the address register.

In the main memory, the complement of the start address is used as data.
Write.

At the start address, a predetermined register RD-REG is read from the main memory.

It is determined whether or not the contents of the register RD-REG are equal to the start address. If not, control is transferred to the process.

If the content of the register RD-REG is equal to the start address, it is normal, so 1 is added to the start address and that is set as a new start address.

It is determined whether or not the start address has reached the predetermined final address. If the final address is not the final address, the control is returned to the processing, and the main memory test is similarly repeated. When the last address is reached, move on to the next test.

If the contents of the registers RD-REG are not equal to the start address, it is an error, so it is determined whether the current mode is the maintenance (CE) mode. In the case of the user mode, the process shifts to the processing and the retry is repeated as in the conventional case. When in the maintenance mode, the process branches to the mini OS and performs a conversation process with an operator or a maintenance person.

The loader 11 shown in FIG.
13 is read out and stored in the main memory 4. In this embodiment, when an error occurs in the process, a process as shown in FIG. 3 is executed. Hereinafter, shown in FIG.
It will be described according to.

The cylinder address of the magnetic disk device 8 in which the OS 13 is stored is set in an address register.

By specifying the cylinder,
Issue a READ instruction.

It is determined whether a read data error has occurred. If an error occurs, the process proceeds to the process. If no error occurs, the process proceeds to the process.

When a read data error occurs, it is determined whether the current mode is the maintenance (CE) mode. If the mode is the maintenance mode, the process branches to the mini OS. In the user mode, a retry is performed as in the conventional case.

When the READ from the magnetic disk device 8 is normally performed, the READ data is set at a predetermined address of the main memory.

 Add 1 to the cylinder address.

Determines whether the cylinder address has reached the last cylinder address, and if not, returns control to the process and similarly reads from the disk.
repeat. When the last cylinder address has been reached, proceed to the next process.

In the present invention, depending on the mode, the mini OS 12
In addition to automatically starting up, the mini OS 12 can be started up at an arbitrary timing by the interrupt switch 17. FIG. 4 shows an example of the interrupt control.

In FIG. 4, reference numeral 21 denotes a flip-flop, 22 denotes an interrupt factor holding register for storing an interrupt factor, 23 denotes an interrupt control circuit, 24 denotes a start address generator, and 25 denotes an interrupt processing program.

When the interrupt switch 17 is operated, the flip-flop
21 is set, and the interrupt factor is stored in the interrupt factor holding register 22. The interrupt control circuit 23 performs control to generate an interrupt according to the state of the interrupt mask or the like. When an interrupt is to be generated, a predetermined start address creating unit 24 is called, a start address of the interrupt processing program 25 is created, and control is passed to the interrupt processing program 25. The interrupt processing program 25 can know the cause of the interrupt by reading the interrupt factor holding register 22. Thus, the interrupt control for activating the mini OS 12 can be realized in the same manner as the conventional interrupt control.

When the mini OS 12 shown in FIG. 1 is started by the initial diagnosis program 10 or the mode determination by the loader 11,
Alternatively, when the operation is started by the operation of the interrupt switch 17, for example, the processing shown in FIG. 5 is executed.

The interrupt factor holding register 22 shown in FIG. 4 is checked to determine whether or not the interrupt switch 17 is ON. If the interrupt switch 17 is ON, the processing is executed.

The interrupt factor held by the interrupt factor holding register 22 is reset.

Display the message "Mini OS startup" on the display.

Enables menu input from the keyboard for conversation processing. For example, in the case of the main memory test, "MM TE
ST [ENTER] ”is input.

 Wait until the [ENTER] key is pressed.

When the [ENTER] key is pressed, the menu is analyzed and executed. For example, in the case of the main memory test, the data in which an error has occurred (the contents of the registers RD-REG) and the start address are displayed.

 Input parameters according to each menu.

 Wait until the [ENTER] key is pressed.

When the [ENTER] key is input, the menu is analyzed and executed according to the input parameters.
Thereafter, the control is returned to the processing, and the conversation processing with the operator is similarly repeated.

Through such conversation processing, the operator can access the CPU, main memory,
A debug function such as read / write to the magnetic disk device can be used.

FIG. 6 shows an example of a display process in the conversation process performed by the mini OS 12, and FIG. 7 shows an example of an input process in the conversation process performed by the mini OS 12.

When displaying menus, etc., as shown in FIG.
PU5 transfers the display contents to the buffer for one screen. The contents of the buffer will be displayed on the display.

In the case of input, the keyboard control unit (not shown)
As shown in FIG. 7, input data from the keyboard is shifted and stored one character at a time in the buffer. [ENTER]
The input is notified to the CPU 5 when a specific key such as a key is input or when the buffer is over.

The commands, menus, and the like prepared in the conversation processing performed by the mini OS 12 may be any as long as they are useful for trouble investigation and debugging, and can be realized by applying well-known technology. Is omitted.

〔The invention's effect〕

As described above, according to the present invention, the maintenance mode is selected even before the OS is started, so that various failure data can be collected and the failure isolation operation of the CMP can be performed at the discretion of the operator or maintenance personnel. Be able to do it. In addition, by selecting the user mode, it is possible to give priority to the start of normal work. In addition, since the mini-OS can be started at an arbitrary timing by an interrupt switch, it can be used for investigating a special failure, initial diagnosis program, initial debugging of a loader, and the like.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of the principle of the present invention, FIG. 2 is an explanatory diagram of a mode judgment in a main memory test according to an embodiment of the present invention, FIG. 4 is an example of interrupt control according to one embodiment of the present invention, FIG. 5 is a processing example of a mini OS according to one embodiment of the present invention, FIG. 6 is an example of display processing according to one embodiment of the present invention, FIG. FIG. 7 shows an example of input processing according to an embodiment of the present invention. In the figure, 1 is CCP, 2 is CCU, 3 is CMP, 4 is main memory, 5 is CPU, 6 is keyboard / display device, 7 is nonvolatile memory, 8 is magnetic disk device, 9 is control circuit, and 10 is control circuit. Initial diagnostics program, 11
Represents a loader, 12 represents a mini OS, 13 represents an OS, 14 represents an application program, 15 represents data, 16 represents a changeover switch, 17 represents an interrupt switch, and 18 represents a mode register.

(72) Inventor Katsuyoshi Okazaki 1015 Uedanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture Inside Fujitsu Limited (72) Inventor Kazuyuki Suzuki 1015, Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture Fujitsu Limited

Claims (1)

(57) [Claims] A non-volatile memory storing a mini-OS having at least an initial diagnosis program, a loader, and a conversation processing function; a changeover switch for switching between a user mode and a maintenance mode; An interrupt switch for instructing an interrupt to be requested, transferring the contents of the non-volatile memory to a main memory at system startup, starting execution from a predetermined address, and executing the initial diagnostic program or the loader. If an error occurs in the user mode, a retry is performed in the user mode, control is passed to the mini OS in the maintenance mode, conversation processing for maintenance is performed, and the operation of the interrupt switch is performed regardless of the user mode or the maintenance mode A method for controlling a processing device, comprising: starting the mini OS and performing maintenance conversation processing.