JPS5933700A - Checking system of memory content - Google Patents

Abstract

PURPOSE:To check easily the storage content of a memory without increasing the memory capacity, by comparing a result of a prescribed operation obtained by reading the storage data on each word region with the reference data to discriminate. CONSTITUTION:A memory 2 to be checked is set in a prescribed state, and at the same time the reference data is set to a register 71 in response to the storage data of the memory 2. Then a CPU1 is started. The storage data of the memory 2 are supplied to the CPU1 and added successively to each other to be stored temporarily. When this addition is over, the address information corresponding to a word region 2N of the memory 2 is delivered. Then the reference data is fetched into the CPU1, compares the result of addition which is stored temporarily with the reference data. The stored data is decided normal if the coincidence is obtained from the comparison and then decided abnormal if no coincidence is obtained. Thus the storage contents of a memory can be easily checked without increasing the memory capacity.

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention relates to a lead/lead used in a microcomputer.
The present invention relates to a memory content checking method that can easily determine whether the contents of a memory such as only memory (ROM) or random access memory (RAM) are correct or incorrect.

[Technical background of the invention]

Recently, the range of use of microcomputers has expanded;;・15
, a wide variety of programs have come to be stored in the program memory. However,
When writing the program code into the program memory, due to various reasons such as a defect in the memory itself, a failure in the writing circuit, or the incorporation of noise, the contents actually written in the memory may not be the contents that should be written. There are different things. First, before or after incorporating a microcomputer into a product, it is necessary to check the contents of the memory.

Therefore, in the past, parity pits and ECC (error detection and correction function) bits were added to each word of the program recorded in the memory, and by checking these bits, it was possible to determine whether the contents stored in the memory were correct or incorrect. .

[Problems with background technology]

However, in such conventional checking methods, since one or more bits of parity pit etc. are added to each word, it is necessary to increase the capacity of the memory. This has the disadvantage that memory is expensive. Furthermore, since the contents of the memory 111 are generally not subject to revision, it is necessary to recognize all the programs stored therein. However, in the above-mentioned conventional checking method, if the stored program itself is correct, the check result will always be iLF, so it could not be determined whether the desired program has been stored completely. Even if a different memory was installed by mistake, this error could not be detected.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide a method for checking the contents of a memory by easily checking the contents of the memory without increasing the memory capacity.

[Keystones of invention]

This invention Q, 1, achieves the above object 7'v, y>,
Among the 7 dominant word areas of the memory, it may be the word area of il-i+, and the memory points 4 to 4 for this word area.
If the data stored in the name word area of the second memory is normal, standard data corresponding to the calculation result of 5N is stored in advance in an external memory having a The storage data of each word order LI area from 11
．． Read out the predetermined above)'1 (calculate the calculation result, and compare this calculation result with the above)', f j%l, data, and from the comparison result L:jC, of each stored data in the memory. To judge whether it is correct or incorrect, set θ to 1m.

[Embodiments of the invention]

FIG. 1 is a schematic configuration diagram of a check device to which a memory content check method is applied in one embodiment of the present invention.
A processor 1 creates a check memory 2 consisting of a check memory 2 and outputs data from the check memory 2 to a data bus 3.
A dirt circuit 4, which reads the address information sent from the upper j+F2 CPU J to the address bus 5, and a decoder circuit 6, which controls opening/closing of the gate circuit 4, and a check circuit 7, are constructed. ing.

The check circuit 7 controls the output of reference data from the base (register 71 for storing all data, setting reference data to this register 71, etc.) switch circuit 72 and the data bus 3. Dirt circuit 7
3, and a decoder circuit 74 which decodes all the address information sent from the CPU 7 to the address bus 5 and controls the opening/closing of the dart circuit 73 and the dart circuit 4 for controlling the output of the checked memory 2 reciprocally. It is composed of. Note that 75 in the figure indicates each of the decoder circuits 6 and 7.
This is an AND circuit that performs logical AND processing on each of the control outputs of 4.

Here, the basic Iff, '+ data stored in the register 71 is for each word negative 1 area 21, . . .
, 2(N-1,') is determined to be the added value when the data stored in N-1,' is normal, and the switch circuit 72 causes the ! ! "!+ will be set in Zosta 7.

On the other hand, the CPU 1 has a check program as shown in the second flowchart, for example, and performs the next check operation according to this program.

That is, the check operation is as follows: (1) 1 . The operation of reading the data stored in each file and performing an addition operation, (11)
The address information corresponding to word shift 2N of the memory to be checked 2 is sent to cause the decoder circuit 74 to open the gate (L path 73), thereby causing the register 71 to also read the reference data 2. Aoi.

Perform a stone operation to compare the results with this reference data.

(iii) As a result of this comparison operation, if the calculation result and the reference data match, the check memo 1 and 2 are
The operation of determining that the stored data is normal, and determining that there is an error in the stored data if they do not match. These are moving breasts displayed on a display section (not shown).

Next, the memory content checking method of this embodiment will be explained based on the above configuration. First, check memory 2
is set in a predetermined state by inserting it into a socket, and the reference data set in advance corresponding to the data stored in the memory 2 to be checked is set in the register 71 by operating the switch circuit 72f.

Then, in this state, CPU 1'lf is started.

Then, the CPU 1 first checks each word area 21. of the memo 1 and 2 to be checked. The address information corresponding to . ~
, 2(N-1)5; Ac-hissed. At this time, the decoder circuit 74 generates an H level signal only when address information of the iQ area 2N arrives.
- R circuit 4 is open because the rf circuit is configured like this.
On the other hand, the gate circuit 73 is in a closed state]
, respectively. Therefore, the data stored in the memory to be checked 2 is inputted to the CPU 1, and the data is stored in the memory 2 to be checked. ,
The data is sequentially processed by the CPU f into three parts.

When this addition 111J is completed, the CPU 1 outputs an address 4 report corresponding to the word area j1ρ, 2N of the memory 2 to be checked. Then, a signal of nH++t level is generated from the decoder circuit 74, and the key-1 circuit 73 is opened in place of the dirt circuit 4. As a result, the reference data set in the Renostaph 1 is transferred to the CPU.
Incorporated into Z. As a result, the CPU 1 compares the previously calculated addition result with the above base Iv (data), and if the two match, the data stored in the memory to be checked 2 is determined to be normal; J of
, "th" is determined to be abnormal and is displayed on the display section. Thus, there is no determination of whether the data stored in the memory 2 is correct or incorrect.

In this way, with the check method of this embodiment, the word area 2Nk of the memory to be checked 2 is used for checking, and the storage capacity is reduced to just 2N words. Since A can be increased in advance and checked by J'i-<;/η by two digits, the check can be performed without increasing the capacity of the memory 2 to the width of Ii8. Therefore, it is possible to reduce the price of each memory.

In addition, in this embodiment, since the reference data can be manually set according to the data stored in the memory to be checked 2, it is possible to manually set the reference data according to the data stored in the memory to be checked. It is possible to reliably detect errors in the A'' class of data stored in the memory, memory errors, etc.

Note that the invention is not limited to the above embodiments (bow:
do not have. For example, in the first embodiment, a case is shown in which base 1" μ data is stored in an external memory (register 71). However, when memo IJ2K data (program) is loaded overnight, The reference data may be input into the word area 2N and checked by reading the reference data from the word area 2N.In this way, at least only the correctness of the stored data itself can be determined.
What 7. (c, .) 1, and an increase in memory capacity can also be prevented. Furthermore, in the above embodiment, 11, the check-only CPo 1 k was used to perform 4 checks, but the CPU actually applied to the product
1, ask for the program for checking JJi L,, this C
The check may be performed while the memory is connected to the PU. In this way, IHI',
, Self as a microcomputer 7.
J-1/bar becomes available source. Also, the performance of each memory data) I
In addition to addition, other calculation methods such as θ; In addition, the number of word areas of the target memory, the number of data points, the volume of sound, the number of data to be calculated, the number of reference data, the circuit configuration, the type of memory, etc. It can be implemented with a small variation of filT/r within a range not departing from the above.

〔Effect of the invention〕

As described above in 1., the present invention allows data to be stored in the word area to be stored in any word area of a plurality of word areas of the memory or in an external memory having a storage capacity U'' for this word area. 51 “JiX Parade Deku Δ” 1]・let L, Tekuki, -hih
Remove the memory data of each word area from its own memory.
1: LII, performs a predetermined operation '('), compares the calculation result with the reference data, and in step 7, determines whether each data stored in the memory is correct or incorrect based on the comparison result.

Therefore, according to the present invention, it is possible to provide a memory content checking method that can easily check the stored content without at least increasing the memory capacity.

[Brief explanation of the drawing]

That is, FIG. 1 is a schematic configuration diagram of a checking device to which a memory content checking method according to an embodiment of the present invention is applied, and FIG. 2 is a flowchart showing an operating program of the CPU of the same device. 1...CPU, 2...Memory to be checked, 3...
・Data bus, 4... Dirt circuit, 5... Address bus, 6... Decoder circuit, 7... Check circuit,
71...Regisf, 72...Switch circuit, 73.
... Keto circuit, 74... Decoder circuit.

Claims (2)

[Claims] (1) If the data stored in each word area of the memory is normal in any word area of a memory that has a storage area for multiple words or in an external memory that has a storage configuration equivalent to this word area, Standard data corresponding to a predetermined calculation result is stored, the stored data is read from each word area of the memory to obtain a calculated value by the predetermined calculation, and this calculated value is compared with the separately read reference data. In-memory check method 0, characterized in that it is configured to determine whether the data stored in the memory is correct or incorrect from the result of the ratio. (2) The external memory is configured such that the reference data can be manually set according to the data stored in the memory.