JPS62249253A - Diagnosis method for content of read only memory - Google Patents

Abstract

PURPOSE:To immediately detect abnormality due to the change of the content of a ROM by performing the arithmetic calculation of a check data for abnormality decision making unnecessary an off-line operation, by utilizing effectively a shared memory. CONSTITUTION:First of al, when a diagnostic program is started up in a state that an initial total sum completion flag 8, and an initial diagnosis completion flag 9 are at off-states, it is decided that a state is an initial state, by processes 11 and 12, and a process 13 is selected,then the arithmetic calculation for an initial total sum is executed. When the diagnostic program is started up at the above state, it is judged that the arithmetic calculation for the initial total sum in an opposite system is completed, by the processes 11 and 12, because only the initial total sum completion flag 8 is set, then a process 15 is selected, and initial diagnosis is executed.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention uses read-only memory (hereinafter referred to as RO) in a two-system system composed of two microprocessors.
RO that detects abnormalities due to changes in the contents of
This relates to a method for diagnosing the contents of M.

[Conventional technology]

FIG. 3 is a flowchart showing a conventional ROM content diagnosis method disclosed in, for example, Japanese Patent Application Laid-Open No. 56-68843. 25 is a process of calculating in units of bytes or words in the order of addresses; 25 is a process of detecting the end of the calculation; and 2G is a process of comparing the calculation result of process 24 with a program for diagnosis.

Next, the operation will be explained. First, the contents of a prespecified number of bytes are calculated off-line, and the calculation results are stored in the ROM as a program for diagnosis. Next, the start address and stop address of the calculation are specified in process 23, and in process 24, it is determined whether the contents of the address order ticRoM have changed from the start address with respect to the final input state (ρ),
Calculations are performed in units of 1 byte or 1 word. When the processing 25 detects that the calculation up to the stop address has been completed, the calculation result is compared with the aforementioned diagnostic program contents stored in the ROM in the processing 26, and is stored in the ROM based on whether or not they match. An abnormality due to a change in the content is detected.

[Problem that the invention seeks to solve]

Since the conventional ROM content diagnosis method is configured as described above, the program for diagnosis as check data for abnormality determination is computed offline and stored in the ROM offline. However, these offline processes require a lot of time, and there is also the possibility of erroneous operations during offline operations, which poses problems such as reducing reliability.

This invention was made to solve the above-mentioned problems, and aims to provide a method for diagnosing the contents of a ROM that can shorten the diagnosis time and improve reliability.

[Means for solving problems]

The ROM content diagnosis method according to the present invention calculates the contents of the ROMs of the two system systems and compares the calculation results through a common memory, and if the results of the comparison match, the calculation results are used for each. The contents of the ROM are calculated periodically in each system, and the calculation results are compared with the unique check data.

[For production]

The ROM content diagnosis method according to the present invention effectively utilizes the shared memory to perform check data calculations for abnormality determination without the need for off-line operations, and to calculate the check data. Immediately detect abnormalities caused by changes.

[Metal practice]

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, 1 is the memory of the A-system microprocessor;
2 is a memory IJ of the B-system threat microprocessor, and 3 is a common memory that can be accessed by both the A and B system microprocessors, and the A and B system microprocessors form a dual system system. 4 and 5 are the ROM area and random access memory (hereinafter referred to as RAM) area of the A-system memory 1;
6゜γ is the ROM area and RAM area of the B system memory 2, the ROM areas 4 and 6 of both the A and B systems are divided into a '' n areas, and the RAM area 5゜ of both the A and B systems. γ is A,
This area is used to store the unique check data for both B systems. 8 is an initial summation completion flag, 9 is an initial diagnosis completion flag, and 10 is a summation data storage area, and a common memory IJ3K is provided for these.

FIG. 2 is a flowchart showing an example of a diagnostic program, in which 11 is a process for checking the initial diagnosis completed fujig 9;
12 is the process of checking initial summation completion 7 lag 8, 13
14 is the process of calculating the initial summation, 14 is the process of turning on the initial summation completion flag 8, 15 is the process of initial diagnosis, 16.2
1 is a process to check whether there is an abnormality, 17 is a process to generate a discrepancy alarm for both systems, 1B is a total data storage area 1
19 is the process of transferring the data of 0 to the RAM areas 5.7 of both systems A and B, and 19 is the process of turning on the initial diagnosis completion flag 9.
20 is a process for periodic diagnosis, and 22 is a process for generating a failure alarm for the own system. This diagnostic program is RO for both A and B systems.
It is stored in a predetermined area in a%n of area M and is activated periodically.

Next, the operation will be explained. First, when the diagnostic program is started with both the initial summation completion flag 8 and the initial diagnosis completion flag 9 off, it is determined that the initial state is reached by the checks in process ii and 12, and process 13 is selected to start the initial summation. The operation is executed. In other words, RO within own system
For each area a%n of M area 4 or 6, the sum of data in 1-byte or 1-word units is calculated in order from the address with the smallest address, and the calculation result is stored in the total data storage area 10 of shared memory 3. It is stored in the corresponding area of a % n. Total calculation result data storage area 1
When the storage to 0 is completed, the initial summation completion flag 8 is turned on in process 14.

When the diagnostic program is started in this state, since only the initial summation completion flag 8 is on, it is determined that the calculation of the initial summation of the partner system has been completed by the check in process 11.12, and process 15 is selected. initial diagnosis is performed. That is, the data in each area a to n of the ROM area 6 or 4 in the own system is calculated in the same manner as in the case described above, and the calculation results are stored in the total data storage area 10 of the common memory 3 by the process 13. It is compared with the computation results of the partner system stored in the ROM, and by detecting a match, an abnormality due to a change in the ROM contents is detected. If a mismatch is detected at this time, it is assumed that a change in content has occurred in either the A-system ROM area 4 or the B-system ROM area 6, and a mismatch alarm for both systems is generated in process 17. If they match completely, the calculation result stored in the total data storage area 10 of the common memory 3 is transferred to the RAM areas 5 and 7 of both systems A and B as unique check data in process 18. . When the transfer of this unique check data is completed, the initial diagnosis completion flag 9 is turned on in process 19.

Thereafter, when this diagnostic program is started at a predetermined period, since the initial diagnosis completion flag 9 is on, the ROM areas 4 and 6 of both systems A and B are checked by the process 11.
Initial diagnosis has been completed, that is, ROM area 4 of both A and B systems.
, 6 is determined to be the final input state, process 20 is selected, and periodic diagnosis is executed within each system of A and B.

That is, the data in ROM areas 4 and 6 are calculated in the same manner as in the above case within each system, and the calculation results are A,
Each of the B systems is compared with its own unique check data transferred to the RAM area 5 or 7, and by detecting a match, an abnormality due to a change in the ROM contents is detected. If an abnormality is detected, a failure alarm for the own system is generated in process 22,
If there is no abnormality, it will wait until it is activated in the next cycle.

〔Effect of the invention〕

As described above, according to the present invention, the ROM content diagnosis method detects abnormalities due to changes in ROM content by performing an initial double-system comparison, and detects abnormalities due to changes in ROM contents using the check data obtained from this initial double-system comparison. Since the configuration is such that periodic KROM content diagnosis is performed for each system, calculation of check data for abnormality detection can also be done by checking the ROM of this check data.
Since all data storage is processed online, wasted time for offline processing can be eliminated, and diagnosis time can be shortened. This ROM content diagnosis method eliminates any factors that reduce reliability, such as erroneous operations caused by offline operations. There is an effect that can be obtained.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of a system to which a ROM content diagnosis method according to an embodiment of the present invention is applied, FIG. 2 is a flowchart showing an example of the diagnosis program, and FIG.
70-Chart for explaining the OM content diagnosis method. 1 is the memory of the A-system microprocessor, 2 is the memory of the B-system microprocessor, 3 is the shared memory, and 4 is the A-system R.
OM area, 5 is 8 system RAM area, 6 is B system ROM area,
7 is a B-system RAM memory, 8 is an initial summation completion flag, 9 is an initial summation completion flag, and 10 is a summation data storage area. Patent applicant Mitsubishi Electric Co., Ltd. Agent Patent attorney 1) Hiroshi Sawa (2 others) −

Claims (1)

[Claims] In a method for diagnosing the contents of a read-only memory having a dual system consisting of two systems of microprocessors storing a program in the read-only memory, a shared memory that can be accessed by both systems of microprocessors is provided; The contents of the read-only memory are calculated in 1-byte or 1-word units in address order,
Comparing the calculation results of both systems through the shared memory,
Through the coincidence detection, an abnormality due to a change in the contents of the read-only memory is detected, and if the calculation results of both systems match as a result of the coincidence detection, the data of the calculation results are transferred to each system. The above-described calculation is periodically performed for each system as check data, and the calculation result is compared with the unique check data, and by detecting a match, it is clarified in which system the abnormality has occurred. A method for diagnosing the contents of read-only memory.