JPS6345645A - Diagnosing system for computer system - Google Patents

Abstract

PURPOSE:To ensure the high-degree inspection of a machine check report function by combining a system diagnosing system with the diagnosing system for the machine check report function and producing a pseudo machine check report at an optional timing. CONSTITUTION:A CPU 2 changes its working environment to diagnose the state of high load while having conflict with plural IOD 91-9n. Under such conditions, the diagnosing routine of a machine check report function is carried out under the fixed conditions to perform the selection of factors of a pseudo machine check error, the designation of generation timing, the storage of the expected value information, burying of the information on the pseudo machine check error, etc. Then the procedure is reset to a main routine for the continuation of the original processing. Then a machine check error generating circuit 5 is started after a designated time by a trigger signal generating circuit 6. Thus a machine check report is produced and the working of this time point is interrupted for check of the propriety and the rightfulness of the report output information. Then the system diagnosis is continued again as long as said output information is normal. If the report output information is not normal, an error is printed and an operator performs a due operation.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a diagnostic method used to verify the validity of a mechanical inspection report function in a computer system.

[Conventional technology]

FIG. 3 is a block diagram of a conventional computer system for diagnosing a machine inspection report function. 3 is a central processing unit (hereinafter referred to as CPU) that executes programs, etc., and 3 is a main memory unit (hereinafter referred to as MMU) that stores programs and data.
, 4 is a diagnostic program (
(hereinafter referred to as program or software), 5
is a mechanical inspection error generation circuit built into C, PU2.

Next, the operation will be explained.

FIG. 4 shows a schematic flowchart of the conventional diagnosis method of the machine inspection report function. In this way, conventionally,
A program created to verify the validity of the machine inspection report function was executed independently to confirm the function when generated statically.

Program 4 stored in MMU 3 in Figure 3 is transferred to CP
This is an ekiben whose operation is defined by successive executions in U2, and the diagnostic method will be explained below. When test program 4 is started, first, after a pseudo machine error occurs,
A fixed area etc. to which the error information is output is initialized (step 20a).

Next, an instruction (hereinafter referred to as Diag) that generates a machine inspection interrupt is given.
Error information, etc. expected from the execution of the command (referred to as a command) is prepared in a predetermined area (step 20b).

Then, a pseudo-machine error is generated. Or execute a Diag instruction to embed the data (step 20C).

There are two types of this instruction: one in which a machine VI inspection interrupt is generated immediately after execution, and the other in which information such as the type of error cause and error occurrence location is embedded in this instruction and generated by the next instruction to be executed (Step 7). 20d). Next, in step 20e, it is checked whether or not a machine inspection error has occurred, and if no machine inspection error has occurred, operator intervention is required. When a simulated machine inspection interrupt occurs, the program compares the actual output information with the expected information (step 20f).
If they match, it is determined whether to continue the next test or not (step 20g); if they do not match, the expected information and execution result information are printed (step 20h), and the process ends with operator intervention.

[Problem that the invention seeks to solve]

In the conventional diagnostic method, as shown in the block diagram in Figure 3 and the diagnostic flowchart in Figure 4, a machine inspection report is generated only when the central processing unit is in a static state, and a machine inspection report is generated only when the central processing unit is in a static state. Because it is a diagnostic method that only takes place inside the body,
Considering the case where the system, such as the user environment, is under a high load and is constantly in a dynamic state, there is a problem in that it cannot be said to be a sufficient diagnostic method to verify this machine inspection reporting function.

This invention was made to solve the above-mentioned problems, and aims to improve the validity of the RAS (Reliability, Availability, and Maintainability) function of the central processing unit, and further improve the quality of computer systems. The aim is to obtain a diagnostic method that can improve the

[Means for solving problems]

The diagnostic method according to the present invention includes a timing generating means for starting the error generating means at an arbitrary timing, and starts the timing generating means after various peripheral devices are started and put into operation by a diagnostic program. The system verifies the validity of the machine inspection report function in an arbitrary operating environment with high load.

[Effect]

In the method according to the present invention, various peripheral devices are first activated by a diagnostic program, and then a trigger signal is generated by activating a timing generation means that generates timing with an arbitrary delay time by the program,
This trigger signal activates the error generating means and generates an interrupt. In other words, from the time the error cause information is embedded until the error report occurs, the central processing unit is operating in competition with multiple peripheral devices, and in this state, machine inspection interrupts due to embedded factors occurs,
Enables validation in dynamic conditions.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an example of the configuration of a computer system in this diagnostic method. In FIG. 3 is the main memory unit (MMU) that stores programs, data, etc.
), 4 is a diagnostic program already stored in the MMU 3, 5 is a mechanical inspection error generation circuit built into the CPU 2,
6 is a trigger signal generation circuit for activating the machine inspection error generation circuit 5; 7 is MMtJ3 and peripheral device control device devices 81 to 7;
channel control device for controlling data transfer with 8n, 91
~9n is the peripheral device control device (hereinafter referred to as ROC)
This is a peripheral device (hereinafter referred to as 10D) controlled by. Note that 5 and 6 provided in the CPU 2 correspond to mechanical inspection error generating means and timing generating means of the present application, respectively.

Next, the operation will be explained based on the block diagram in FIG. 1 and the diagnostic flowchart in FIG. 2.

First, the CPU 2 starts system diagnosis using the program 4 stored in the MMU 3. The system diagnostic program activates all l0Ds 91 to 9n in order to start their operations (step lQa,
10b). The activated l0D91~9n are each MM
The operation starts according to the test routine based on the data transfer between U3 and l0C81~8n (steps 10c and 10).
d). On the other hand, when all l0Ds 91 to 9n start operating, the CPU 2 starts self-diagnosis (step 10e).
. After that, when a report of operation completion or abnormal termination is received from the operating rODs 91 to 9n (step 10f),
The CPU 2 interrupts the child strike, processes the report (step 10g), and then interrupts the stopped child strike 10D91 to 9n.
and continues the child strike again (step 10a).
~10d) In this way, CPU2 uses multiple l0D9
While competing with 1 to 9n, the operating environment is changed and diagnosis is executed under high load conditions. In this state, the diagnostic routine of the machine inspection report function (■ in step 10e) is executed under certain conditions, and the timing of occurrence of Selection 2 of the pseudo machine inspection error factor is specified.

After processing such as storing expected value information and embedding pseudo machine test error information, the program returns to the main routine and continues the original processing. Thereafter, the mechanical inspection error generating circuit 5 is activated by the trigger signal generating circuit 6 after a specified time has elapsed, and a mechanical inspection report is generated. When a machine inspection report occurs (Step 1
0h), the operation at that point is interrupted, check the validity and validity of the report output information (Step 10m), and if normal, continue the system diagnosis again (Step 10e).
If there is an abnormality, that is, a mismatch, an error will be printed in one step and the operator will have to intervene. Incidentally, in step 10i, it is determined whether or not the machine inspection report diagnosis routine indicated by ■ in step 10e is being executed. If the machine inspection report function is being executed, the diagnosis of the machine inspection report function as described above is performed, and if it is not being executed, a true machine inspection error report has occurred. Error handling occurs.

In this embodiment, the validity of the machine inspection report function during system diagnosis can be verified by changing the type of occurrence factor through program intervention and causing it to occur after an arbitrary period of time has elapsed.

〔Effect of the invention〕

As described above, according to the present invention, the system diagnosis method is combined with the diagnosis method of the machine inspection report function, and the simulated machine inspection report is generated at an arbitrary timing to perform advanced verification of the machine inspection report function. This has the effect of improving the reliability and quality of the computer system.

[Brief explanation of drawings]

FIG. 1 is a system configuration block diagram showing an embodiment of the present invention, FIG. 2 is a flowchart showing the operation of the embodiment, and FIG. 3 is a system configuration block diagram showing a conventional diagnostic method.
FIG. 4 is a schematic flowchart of a conventional diagnostic method. ■...Computer system main body, 2...Central processing unit,
3...Main storage device, 4...Diagnostic program, 5...
・Mechanical inspection error generation circuit (mechanical inspection error generation means), 6
...Trigger signal generation circuit (timing generation means), 9
1~9n...Peripheral equipment. In addition, in the figures, the same reference numerals indicate the same or corresponding parts.

Claims (1)

[Claims] A computer system with a machine inspection report function is equipped with a machine inspection error generation means, and by running a diagnostic program and generating a pseudo machine error, the expected value and the actual value of the machine inspection report are compared, and the machine inspection report function is In a computer system diagnostic method designed to verify validity, a timing generating means is provided which activates the above-mentioned error generating means at an arbitrary timing, and after starting various peripheral devices and bringing them into operation state by a diagnostic program, the above-mentioned error generating means is activated. A computer system diagnosis method characterized in that the validity of a mechanical inspection report function is verified in an arbitrary operating environment where the system is subject to high load by activating a timing generation means.