JPH10269147A - Data checking method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data checking method that detects the switching of data blocks when the data blocks used for an operation are switched, also detects data zero when entire data become zero because of a failure, etc., and further detects the inversion of bits when the bits of an even data block are inserted. SOLUTION: A controlling means divides data with a prescribed length and allocates an address to each of plural data blocks. Next, it takes exclusive-OR of each value of the data block and a corresponding address and further takes the total sum of acquired results. Next, it takes exclusive-OR of the complement of two of the total sum and the address of the check word and further creates a check word by adding a preliminarily defined value to the resultant value of the exclusive-OR. Then, the check word created in this way is written in a storage means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a processing method of means for verifying the validity of information developed on a memory of an information processing device and a communication processing device on a block basis.

[0002]

2. Description of the Related Art As a means for easily verifying the validity of information developed on a memory of an information processing device and a communication processing device, a method using a checksum and a method using a horizontal parity check are known.

FIG. 1 is a configuration diagram showing a configuration for realizing a conventional data check. In FIG. 1, reference numeral 1 denotes control means which is a main part for performing a data check method, and is constituted by a CPU or the like. Reference numeral 2 denotes a storage unit such as a memory.

FIG. 18 is an explanatory diagram showing an example of a conventional data check method using a checksum. In the method using the checksum, as shown in FIG. 18, data of a block to be checked is divided into a certain unit length (for example, 2 bytes) and divided into n calculation units. Let these data be data [k] (0 ≦ k ≦ n−1). Also, the position where the check word is stored is j, and data
It is assumed that a check word is stored in [j].

At this time, according to the checksum method, if the following equation is satisfied, it is determined that the data in the block is valid.

[0006]

(Equation 1)

In this equation, a is a constant such as 0.

[0008] When updating data at an arbitrary position k to new data data '[k] in a block whose data is guaranteed to be valid by the data check method, a value data' [ j] is updated. (1) A value data ″ [k] is calculated by adding 1 to a value obtained by inverting all the bits of the new data data ′ [k]. In the following equations, all bit inversion (one's complement) operation is described as “inversion”.

[0009]

(Equation 2)

(2) The calculation result of the above (1), data data [k] to be changed, and the check word d before the change
The sum of data [j] is calculated and used as the changed check word.

[0011]

(Equation 3)

FIG. 19 is an explanatory view showing another example of a conventional data check method using a checksum. FIG.
As shown in FIG. 9, the data of the block to be checked is divided into a certain unit length (for example, 2 bytes) and divided into n calculation units. These data are stored in data [k] (0
≤ k ≤ n-1). The check word is stored outside the area of the block to be checked, and is set to checkword here.

In such a conventional data check method using a checksum, data in a block may be determined to be valid when the following equation is satisfied.

[0014]

(Equation 4)

Further, when updating data at an arbitrary position k in a block in which data is guaranteed to be valid by a conventional data check method using a checksum, all bits of data data [k] before change are updated. The value obtained by adding 1 to the value obtained by inverting the value, the new data data '[k], and the sum of the check word checkword before change and checkword
The check word is updated to d 'and the new data data'
The data is updated to [k].

[0016]

(Equation 5)

FIG. 20 is an explanatory diagram showing an example of a conventional data check method using a horizontal parity check. As shown in FIG. 20, the data of the block to be checked is divided into unit lengths (for example, 2 bytes).
Divide into n calculation units. Data of those data
[K] (0 ≦ k ≦ n−1). The position where the check word is stored is j, and the check word is stored in data [j].

In a conventional data check method using a horizontal parity check, data in a block is determined to be valid when the following equation is satisfied. In the following expressions, the exclusive OR operation is described by a symbol “記号”.

[0019]

(Equation 6)

In this equation, a is a constant, for example, a
= A5A5h (where h indicates a hexadecimal number) or the like.

When initializing data in a block in a conventional data check method using a horizontal parity check, a check word data [j] corresponding to the data is obtained by the following equation.

[0022]

(Equation 7)

When updating data at an arbitrary position k in a block in which data is guaranteed to be valid by a conventional data check method using a horizontal parity check, the value data '[ j] and the data is updated to new data data '[k].

[0024]

(Equation 8)

FIG. 21 is an explanatory diagram showing another example of a conventional data check method using a horizontal parity check. As shown in FIG. 21, the data of the block to be checked is divided into unit lengths (for example, 2 bytes).
Divide into n calculation units. Data of those data
[K] (0 ≦ k ≦ n−1). The check word is stored outside the area of the block to be checked, and is set to checkword here.

In this conventional example, when the following equation is satisfied, it is determined that the data in the block is valid.

[0027]

(Equation 9)

When updating data at an arbitrary position k in a block in which data is guaranteed to be valid by the data check method in the conventional example, a check word 'obtained by the following equation is added to a check word. Is updated to new data data '[k].

[0029]

(Equation 10)

[0030]

A conventional method using a checksum and a data check method using a horizontal parity check are configured as described above, and the calculation used for the checksum calculation and the horizontal parity check calculation is performed as follows. This is an operation for which the commutation rule holds. For this reason, when the data blocks used for the operation are switched, the operation result of the check word is unchanged in the conventional method.
There has been a problem that data interchange cannot be detected.

In the data check method using the conventional checksum, the check word becomes 0 when all data is 0.
A situation in which the data all become 0 unintentionally cannot be detected, and even in this case, there is a problem that the data is erroneously determined to be valid.

In the data check method using the conventional horizontal parity check, when the bits of the even-numbered data block are inverted, the same operation result is obtained as when the exclusive OR is not used. There was a problem that it could not be detected.

The present invention has been made to solve such a problem, and when data blocks used for calculation are replaced, the replacement of the data is detected, and the case where all data are 0 is detected. It is an object of the present invention to provide a data check method for detecting and inverting a bit of an even-numbered data block when the bit is inverted.

[0034]

A data check method according to a first aspect of the present invention includes a data dividing step of dividing data into predetermined lengths to generate a plurality of first data blocks;
An address allocating step of allocating an address to each of the plurality of first data blocks, and exclusive-use of an address corresponding to the value of the first data block to each of the plurality of first data blocks. A second data block creation step of creating a plurality of second data blocks by taking a logical sum; and for checking data reliability,
A first check word insertion step of inserting or adding a first check word to a predetermined address within an address range of the first data block, and a summation of the plurality of second data blocks, The two's complement of the sum of the two data blocks is calculated, then the exclusive OR of the two's complement of the sum and the address of the first check word is calculated, and the result of the exclusive OR is calculated. A first check word creating step of creating a first check word by adding a predetermined value to a value; and a writing step of writing the first data block and the first check word into storage means. And reading out the plurality of first data blocks and the first check word from the storage means, for each of the plurality of first data blocks, An exclusive OR operation value of Taburokku and corresponding address, and a third data block creation step of creating a plurality of third data blocks, the first
XORing the value of the check word and the corresponding address to create a second check word, and all of the plurality of third data blocks and the second check word. A fourth data block creating step of creating a fourth data block by summing the check data with a check word; and a checking step of determining whether or not the fourth data block is the same as the predetermined value. Things.

Further, in the data check method according to the second invention, when data is updated, an exclusive OR of the value of the first check word before the update and the corresponding address is obtained, and A first intermediate data creating step of creating intermediate data, and a first data block to be changed among a plurality of first data blocks, the value of which is associated with an address corresponding to the value of the first data block. A second intermediate data creating step of taking an exclusive OR to create a plurality of second intermediate data; and taking an exclusive OR of the changed first data block and the address of the first check word. A third intermediate data creating step of creating third intermediate data; and taking a sum of the first intermediate data, the plurality of second intermediate data, and the two's complement of the third intermediate data. Sum and above Is obtained by a step of creating a first check word after the data updating takes the exclusive OR of the address of one of the check words.

The data checking method according to a third aspect of the present invention includes a data dividing step of dividing data into predetermined lengths to generate a plurality of first data blocks, An address allocating step of allocating an address to each of the plurality of first data blocks; a corresponding address is added to the value of the first data block to form a plurality of second data blocks; A second data block creating step of creating, and a first check word inserting step of inserting or adding a first check word to a predetermined address within the address range of the first data block for data reliability check And a third data block creation step of taking the exclusive OR of all of the plurality of second data blocks to create a third data block XORing the value of the third data block with a predetermined value to create a fourth data block, and a fourth data block creating step of creating a fourth data block; Take the two's complement and add the first data block to the fourth data block.
A writing step of writing the first data block and the first check word into storage means; and a writing step of writing the first data block and the first check word into storage means.
When reading the data block and the first check word from the storage means, for each of the plurality of first data blocks, add a corresponding address to the value of the first data block. A fifth data block creating step of creating a plurality of fifth data blocks; and a second check word creating step of adding a corresponding address to the value of the first check word to create a second check word. A sixth data block creating step of taking an exclusive OR of all of the plurality of fifth data blocks and the second check word to create a sixth data block; And a check step of determining whether the value is the same as the predetermined value.

In the data check method according to the fourth invention, when data is updated, a corresponding address is added to the value of the first check word before the update, and the first check word is updated.
A first intermediate data creating step of creating intermediate data of a first data block and a first intermediate data to be changed among a plurality of first data blocks.
A second intermediate data creating step of adding a corresponding address to the value of the first data block to create a second intermediate data for the data block of the first data block;
Add the corresponding address to the value of the data block,
A third intermediate data creating step of creating third intermediate data, and an exclusive OR of the first intermediate data, the second intermediate data, and the third intermediate data to obtain fourth intermediate data And a step of adding a two's complement of the address of the first check word to the fourth intermediate data to create a check word after data update. is there.

The data checking method according to a fifth aspect of the present invention includes a data dividing step of dividing data into predetermined lengths to create a plurality of first data blocks; An address allocating step of allocating an address to each of the plurality of first data blocks; and calculating, by exclusive ORing, for each of the plurality of first data blocks, a value of the first data block and an address corresponding to the plurality of second data blocks; A second data block creating step of creating the first data block, a first data block storing step of storing the plurality of first data blocks in the first storage means, and a sum of the plurality of second data blocks. A check word creating step of creating a check word based on the obtained sum, and a check word creating step of storing the check word in the second storage means. A storing step, and when reading the plurality of first data blocks from the first storage means, for each of the first data blocks, a value of the first data block and an address corresponding to the value are stored. A third data block creation step of taking an exclusive OR to create a plurality of third data blocks, and a fourth data block creation of taking a sum of the plurality of third data blocks to create a fourth data block And a check step of determining whether or not the fourth data block is the same as the check word read from the second storage means.

Further, in the data checking method according to the sixth invention, when the data is updated, an exclusive OR of the changed first data block and the corresponding address is obtained, and the first intermediate data is updated. A first intermediate data creating step of creating a first intermediate data block and a first intermediate data block of the plurality of first data blocks before the change which is a target of the change.
XORing the value of the data block with the corresponding address to create the second intermediate data, the second intermediate data creating step, and the second intermediate data of the first intermediate data and the second intermediate data. And the step of taking the sum of the complement of the check word and the value of the check word before update to create a check word after data update.

The data checking method according to a seventh aspect of the present invention includes a data dividing step of dividing the data into predetermined lengths to create a plurality of first data blocks; An address allocating step of allocating an address to each of the plurality of first data blocks; a corresponding address is added to the value of the first data block to form a plurality of second data blocks; A second data block creating step of creating, a first data block storing step of storing the plurality of first data blocks in the first storage means, and an exclusive OR between the plurality of second data blocks. Then, a check word creating step of creating a check word based on the result of the exclusive OR, and storing the check word in the second storage means. And when reading the plurality of first data blocks from the first storage means, for each of the first data blocks, an address corresponding to the value of the first data block. A third data block creating step of creating a plurality of third data blocks by adding a third data block, and a fourth data block of creating the fourth data block by taking an exclusive OR of all the plurality of third data blocks. The method includes a data block creation step and a check step of determining whether or not the fourth data block is the same as the check word read from the second storage means.

Further, in the data checking method according to the eighth invention, when the data is updated, the first data block to be changed among the plurality of first data blocks is subjected to the first data block. A first intermediate data creating step of creating a plurality of first intermediate data by adding a corresponding address to the value of the data block, and adding the corresponding address to the changed value of the first data block. A second intermediate data creating step of creating second intermediate data, an exclusive OR of the check word value before update, the first intermediate data, and the second intermediate data is calculated. And creating a check word after updating the data.

Further, in the data checking method according to the ninth invention, when the checking step reads the first data block and the first check word from the storage means, each value of the first data block is read. Are all 0,
The exclusive OR of the sum of the plurality of third data blocks created by taking the exclusive OR of the value of the first data block and the corresponding address, and the exclusive check of the address corresponding to the first check word It is to judge whether or not the sum of the second check word created by taking the sum is the same as a predetermined value.

Also, in the data checking method according to the tenth aspect, when the checking step reads out the first data block and the first check word from the storage means, each value of the first data block is read. Are all 0, all of the plurality of third data blocks created by adding the corresponding address to the value of the first data block and the address corresponding to the first check word, respectively. To determine whether or not the exclusive OR of the second check word created by adding .gamma. Is equal to a predetermined value.

[0044]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 FIG. FIG. 1 is a configuration diagram showing a configuration for realizing a data check according to the present invention. In FIG. 1, reference numeral 1 denotes control means which is a main part for performing a data check method, and is constituted by a CPU or the like. Reference numeral 2 denotes a storage unit such as a memory.

FIG. 2 is an explanatory diagram showing an embodiment of the data check method according to the present invention. Next, FIG.
The operation of the data check method shown in FIG. First, the data of the block to be checked is divided into a certain unit length (for example, 2 bytes) and divided into n calculation units. These data are converted to data [k] (0 ≦ k ≦ n−
1). Further, a sequence indicating the relative position of each calculation unit data [k] in the block is represented by num [k] (0 ≦
k ≦ n−1).

It is assumed that the cycle of the sequence num [k] is equal to or greater than the number n of the divided calculation units. In this embodiment, for example, a sequence of num [k] = k is used as num [k].
Other sequences may be used. The position where the check word is stored is j, and the check word is stored in data [j].

At this time, in the first embodiment, if the following equation is satisfied, it is determined that the data in the block is valid. In the following equation, the exclusive OR operation is described with a symbol “^”.

[0048]

[Equation 11]

In this equation, a is a constant.
= 0.

In the first embodiment, when data in a data block is initialized, a corresponding check word data [j] is obtained by the following equation.

[0051]

(Equation 12)

When updating data at an arbitrary position k in a block whose data is guaranteed to be valid by the data check method according to the first embodiment, the value data ′ [ j] and the data may be updated to new data data '[k].

[0053]

(Equation 13)

In the first embodiment, the case where the exchange of the k-th data and the m-th data cannot be detected is limited to the case where the following equation is satisfied.

[0055]

[Equation 14]

Next, specific examples will be described below. For example, in the initialization processing, the data is set to “000010110”.
A description will be given of the case of setting 0101010 ″.

When this data is divided by 3 bytes, the following result is obtained: 0th data = 000 1st data = 011 1st data = 010 3rd data = 101 4th data = 010

In this case, if the position of the check word is j = 5, for example, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100

Therefore, in this case, Becomes

Accordingly, the check word is data [5] = [000+ {inversion (110 + 000) +1}]｝ 101 = {000+ (001 + 1)｝ ＾ 101 = (000 + 010) ＾ 101 = 111.

Therefore, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 111 ... num [5] = 101 ... checkword

Therefore, in this case, data

[0] @num

[0] = 000 data [1] @num [1] = 010 data [2] @num [2] = 000 data [3] @num [3] = 110 data [4] @num [4] = 110 data [5] $ num [5] = 010 ... Check word a = 000.

In this case, when aforesaid data is read out, since a = 000, the apparatus determines that this data is valid.

Next, the operation when data at an arbitrary position is changed in this data block will be described. For example, the data data [2] of num [2] is set to "01".
When changing from “0” to “000”, the changed dat
a ′ [2] = 000.

Therefore, in this case, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data '[2] = 000... Num [2] = 010... Data after change data [3] = 101. 3] = 011 data [4] = 010... Num [4] = 100

Therefore, in this case, data

[0] $ num

[0] = 000 data [1] ＾ num [1] = 010 data '[2] ＾ num [2] = 010... Data after change data [3] ＾ num [3] = 110 data [4] 変 更num [4] = 110 Total = 000

Therefore, the new check word is: data [5] = [010 + 000 + {inversion (000 ＾ 010) +1}] ＾ 10 1 = {010+ (101 + 1)｝ ＾ 101 = 000 ＾ 101 = 101

Therefore, data

[0] = 000… num

[0] = 000 data [1] = 011 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 101 ... num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 101 ... num [5] = 101 ... new check word

Therefore, in this case, data

[0] @num

[0] = 000 data [1] @num [1] = 010 data [2] @num [2] = 010 data [3] @num [3] = 110 data [4] @num [4] = 110 data [5] $ num [5] = 000 ... New check word a = 000.

As described above, when the data is changed, a new check word is not obtained from all data, but
An extremely simple calculation can be performed based on the data to be changed, and the calculation can be speeded up.

Next, the operation in the case where data exchange has occurred will be described. For example, when the 0th data and the 2nd data are exchanged, the data before the data exchange is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 111 ... num [5] = 101 ... checkword

The data

[0] @num

[0] = 000 data [1] @num [1] = 010 data [2] @num [2] = 000 data [3] @num [3] = 110 data [4] @num [4] = 110 data [5] $ num [5] = 010 ... Check word a = 000.

On the other hand, the data after the data exchange is data

[0] = 010 ... num

[0] = 000 ... Data after replacement data [1] = 011 ... num [1] = 001 data [2] = 000 ... num [2] = 010 ... Data after replacement data [3] = 101... Num [3] = 011 data [4] = 010... Num [4] = 100 data [5] = 111... Num [5] = 101.

Therefore, data

[0] @num

[0] = 010 ... Data after replacement data [1] @num [1] = 010 data [2] @num [2] = 010 ... Data after replacement data [3] @num [3] = 110 data [4] @num [4] = 110 data [5] @num [5] = 010 ... check word a = 100.

As described above, for example, when the storage position of the data in the data block is changed due to a malfunction of the device, the value of a, that is, the sum changes. Of the data storage position can be detected, and countermeasures can be taken.

The operation of the control means 1 shown in FIG. 1 in this embodiment will be described below with reference to a flowchart. FIG. 3 is a flowchart showing the data initialization operation of the control means 1 in this embodiment. Next, the operation of data initialization will be described using the flowchart operation shown in FIG.

First, the control means 1 divides data to create a plurality of data blocks (step S301). Next, the control means 1 assigns a different address to each data block (step S302). Next, the control unit 1 creates a plurality of second data blocks by an exclusive OR of each data block and the corresponding address (step S303).

Next, the control means 1 creates a check word from the sum of the second data block and the address of the check word in accordance with equation (12) (step S30).
4). Next, the control means 1 inserts the above-mentioned check word into the data block (step S305). Next, the control means 1 writes the data into the storage means 2 (step S
306).

FIG. 4 is a flowchart showing a data check operation of the control means 1 in this embodiment. Next, the data check operation will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data by the same method as at initialization to create a plurality of data blocks, and assigns different addresses to each data block (step S401). Next, the control means 1 creates a third data block by the exclusive OR of the data block and the address (step S402). Next, the control means 1 creates a second check word by the exclusive OR of the check word and the address (step S403).

Next, the control means 1 creates a fourth data block by calculating the sum of all the third data blocks and the second check word (step S40).
4). Next, the control means 1 compares the fourth data block with a predetermined value (step S405), and if they match, determines that the data is normal (step S40).
6). If the comparisons in step S405 do not match, it is determined that the data is abnormal (step S4).
07).

FIG. 5 is a flow chart showing the operation of the data updating process of the control means 1 in this embodiment. Next, the operation of the data update process will be described using the flowchart operation shown in FIG.

First, the control means 1 creates a plurality of data blocks by dividing data in the same manner as in the initialization, and assigns different addresses to each data block (step S501). Next, the control means 1 creates first intermediate data by exclusive ORing the check word before data update and the address corresponding to the check word (step S502). Next, the control means 1 creates the second intermediate data by the exclusive OR of the data block before the data update and the address corresponding to the data block (step S503).

Next, the control means 1 creates third intermediate data by exclusive-ORing the address corresponding to the updated data block with the data block (step S504). Based on the first to third intermediate data and the address of the check word, an updated check word is created according to equation (13) (step S505). Next, the control means 1 writes the updated data block and check word into the storage means 2 (step S506).

Embodiment 2 FIG. 6 is an explanatory diagram showing another embodiment of the data check method according to the present invention. Next, the data check method shown in FIG. 6 will be described. As shown in FIG. 6, the data of the block to be checked is divided into unit lengths (for example, 2 bytes).
Divide into n calculation units. Data of those data
[K] (0 ≦ k ≦ n−1).

A sequence indicating the relative position of each calculation unit data [k] in the block is represented by num [k] (0 ≦
k ≦ n−1). If one cycle of the sequence num [k] is equal to or more than the length of a calculation unit, for example, a number that can be represented by 2 bytes, and is equal to or less than the number of calculation units, the inversion of the data order can be detected. Num [k] = k
However, another sequence may be used. The position where the check word is stored is j, and the check word is stored in data [j].

In the second embodiment, if the following equation is satisfied, it is determined that the data in the block is valid.

[0088]

(Equation 15)

In this equation, a is a constant.
= A5A5h.

In the second embodiment, when data in a block is initialized, a corresponding check word data [j] is obtained by the following equation.

[0091]

(Equation 16)

When updating data at an arbitrary position k in a block whose data is guaranteed to be valid by the data check method according to the second embodiment, the value data ′ [ j] and the data may be updated to new data data '[k].

[0093]

[Equation 17]

In the second embodiment, when the switching of the k-th data and the m-th data cannot be detected, it is limited to the case where the following equation is satisfied.

[0095]

(Equation 18)

Next, specific examples are shown below. For example, a case will be described where, in the initialization processing, data is set to “000011010101010”, which is the same as in the first embodiment, and a check word is determined so that each bit becomes an even number in the horizontal parity check. In this case, the horizontal parity a is set to 010.

When this data is divided into 3 bytes, the following is obtained: 0th data = 000 1st data = 011 1st data = 010 3rd data = 101 4th data = 010

In this case, if the position of the check word is j = 5, for example, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100

Therefore, in this case, Becomes

Therefore, the check word is data [5] = {010/000/100/100/000/110} + {inversion 101 + 1} = 010/110 + (010 + 1) = 100 + 011 = 111

Therefore, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 111 ... num [5] = 101 ... checkword

Therefore, in this case, data

[0] + num

[0] = 000 data [1] + num [1] = 100 data [2] + num [2] = 100 data [3] + num [3] = 000 data [4] + num [4] = 110 data [5] + num [5] = 100... Check word a = 010.

In this case, when the data is read out, if a = 010, the apparatus determines that the data is valid.

Next, the operation when data at an arbitrary position is changed in this data block will be described. For example, the data data [2] of num [2] is changed to “01”.
When changing from “0” to “000”, the changed dat
a ′ [2] = 000.

Therefore, in this case, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data '[2] = 000... Num [2] = 010... Data after change data [3] = 101. 3] = 011 data [4] = 010... Num [4] = 100

Therefore, in this case, data

[0] + num

[0] = 000 data [1] + num [1] = 100 data '[2] + num [2] = 010 ... changed data data [3] + num [3] = 000 data [4] + num [4] = 110 EXOR = 000.

Accordingly, the new check word is data [5] = {100 {010} 100} + {inversion 101 + 1} = 010 + (010 + 1) = 010 + 011 = 101.

Therefore, data

[0] = 000… num

[0] = 000 data [1] = 011 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 101 ... num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 101 ... num [5] = 101 ... new check word

Therefore, in this case, data

[0] + num

[0] = 000 data [1] + num [1] = 100 data [2] + num [2] = 010 data [3] + num [3] = 000 data [4] + num [4] = 110 data [5] + num [5] = 010 ... new check word a = 010

As described above, when data is changed, a new check word is not obtained from all data, but
An extremely simple calculation can be performed based on the data to be changed, and the calculation can be speeded up.

Next, the operation in the case where data exchange has occurred will be described. For example, when the 0th data and the 2nd data are exchanged, the data before the data exchange is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 111 ... num [5] = 101 ... checkword

[0112] data

[0] + num

[0] = 000 data [1] + num [1] = 100 data [2] + num [2] = 100 data [3] + num [3] = 000 data [4] + num [4] = 110 data [5] + num [5] = 100 The check word a = 010.

On the other hand, the data after the data exchange is data

[0] = 010 ... num

[0] = 000 data [1] = 011 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 101 ... num [3] = 011 data [4 ] = 010 ... num [4] = 100 data [5] = 111 ... num [5] = 101 ... check word. Note that the check word is the same as before.

Therefore, data

[0] + num

[0] = 010 data [1] + num [1] = 100 data [2] + num [2] = 010 data [3] + num [3] = 000 data [4] + num [4] = 110 data [5] + num [5] = 100... Check word a = 110.

As described above, for example, when the data storage position in the data block is switched due to a malfunction of the device, the value of a, that is, the exclusive OR (EXO)
Since the value of the horizontal parity of the value obtained by R) changes, by examining the value of a, it is possible to detect the change of the storage position of the data in the data block, and to take measures.

The operation of the control means 1 shown in FIG. 1 in this embodiment will be described below with reference to a flowchart. FIG. 7 is a flowchart showing the data initialization operation of the control means 1 in this embodiment. Next, the operation of data initialization will be described using the flowchart operation shown in FIG.

First, the control means 1 divides data to create a plurality of data blocks (step S701). Next, the control means 1 assigns a different address to each data block (step S702). Next, the control unit 1 creates a plurality of second data blocks by adding a corresponding address to each data block (step S703). Next, the control means 1 creates a third data block by performing an exclusive OR operation on all of the plurality of second data blocks (step S704).

Next, the control means 1 performs an exclusive OR operation on the third data block and a predetermined value to create a fourth data block (step S705). Next, the control means 1 creates a check word from the address of the fourth data block and the check word according to equation (16) (step S706). Next, the control means 1 inserts the check word into the data block (step S707). Next, the control unit 1 writes the data including the check word and the data block into the storage unit 2 (step S708).

FIG. 8 is a flowchart showing the data check operation of the control means 1 in this embodiment. Next, the data check operation will be described using the flowchart operation shown in FIG.

First, the control means 1 divides data by the same method as at the time of initialization to create a plurality of data blocks, and assigns different addresses to each data block (step S801). Next, the control means 1 assigns
A plurality of fifth data blocks are created by adding the corresponding addresses (step S802). Next, the control means 1 creates a second check word by adding the corresponding address to the check word (step S803).

Next, the control means 1 creates the sixth data block by taking the exclusive OR of all of the fifth data block and the second check word (step S804). Next, the control means 1 compares the sixth data block with a predetermined value (step S805), and if they match, determines that the data is normal (step S8).
06). If the comparisons in step S805 do not match, it is determined that the data is abnormal (step S805).
807).

FIG. 9 is a flow chart showing the operation of the data updating process of the control means 1 in this embodiment. Next, the operation of the data update process will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data by the same method as at the time of initialization to create a plurality of data blocks, and assigns different addresses to each data block (step S901). Next, the control unit 1 creates first intermediate data by adding an address corresponding to the check word before data update (step S902). Next, the control unit 1 creates the second intermediate data by adding the address corresponding to the data block before the data update (step S903).

Next, the control means 1 creates third intermediate data by adding an address corresponding to the updated data block (step S904). Next, the control means 1 creates an updated check word based on the equation (17) based on the first to third intermediate data and the address of the check word (step S905).
Next, the control unit 1 writes the updated data block and check word into the storage unit 2 (step S906).

Embodiment 3 FIG. 10 is an explanatory diagram showing another embodiment of the data check method according to the present invention. As shown in FIG. 10, the data of the block to be checked is divided into unit lengths (for example, 2 bytes).
Divide into n calculation units. Data of those data
[K] (0 ≦ k ≦ n−1). Also, each calculation unit d
A sequence indicating the relative position of ata [k] in the block is defined as num [k] (0 ≦ k ≦ n−1).

It is assumed that the period of the sequence num [k] is equal to or greater than the number n of the divided calculation units. Further, in this embodiment, for example, a sequence of num [k] = k is used as num [k], but another sequence may be used. The check word is stored outside the area of the block to be checked, and is set to checkword here.

At this time, in the third embodiment, if the following equation is satisfied, it is determined that the data in the block is valid.

[0128]

[Equation 19]

When updating data at an arbitrary position k in a block whose data is guaranteed to be valid by the data check method according to the third embodiment, the value checkword 'obtained by the following equation is used. The check word may be updated, and the data may be updated to new data data '[k].

[0130]

(Equation 20)

Next, specific examples will be described below. For example, a case will be described where, in the initialization processing, the data is set to “00001101010101010” as in the first embodiment, and the sum of the data is used as a check word.

When this data is divided by 3 bytes, the following result is obtained: 0th data = 000 1st data = 011 1st data = 010 3rd data = 101 4th data = 010

In this case, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100

Therefore, in this case, data

[0] @num

[0] = 000 data [1] @num [1] = 010 data [2] @num [2] = 000 data [3] @num [3] = 110 data [4] @num [4] = 110 Become. Therefore, the check word that is the lower 3 bits of the sum of these data is check word = 110.

Next, an operation when data at an arbitrary position is changed in this data block will be described. For example, the data data [2] of num [2] is changed to “01”.
When changing from “0” to “000”, the changed dat
a ′ [2] = 000.

The data before the change is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010... Num [4] = 100.

Therefore, in this case, data

[0] @num

[0] = 000 data [1] @num [1] = 010 data [2] @num [2] = 000 data [3] @num [3] = 110 data [4] @num [4] = 110 Become. Therefore, the check word that is the lower 3 bits of the sum of these data is check word = 110.

The data after the change is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data '[2] = 000... Num [2] = 010... Data after change data [3] = 101. 3] = 011 data [4] = 010... Num [4] = 100

Therefore, in this case, data

[0] $ num

[0] = 000 data [1] ＾ num [1] = 010 data '[2] ＾ num [2] = 010... Data after change data [3] ＾ num [3] = 110 data [4] 変 更num [4] = 110. Therefore, the check word that is the lower 3 bits of the sum of these data is check word = 000.

In this embodiment, the check word after the data change is calculated by the following operation. Check word after change = 010 + (inversion 000 + 1) + 110 = 010 + (111 + 1) + 110 = 010 + 000 + 110 = 000

As described above, when data is changed, a new check word is not obtained from all data,
An extremely simple calculation can be performed based on the data to be changed, and the calculation can be speeded up.

Next, the operation when data exchange is performed will be described. For example, when the 0th data and the 2nd data are exchanged, the data before the data exchange is data

[0] = 000… num

[0] = 000 ... Data to be replaced data [1] = 011 ... num [1] = 001 data [2] = 010 ... num [2] = 010 ... Data to be replaced data [3] = 101 ... num [3] = 011 data [4] = 010 ... num [4] = 100

Data

[0] @num

[0] = 000 ... Data to be replaced data [1] @num [1] = 010 data [2] @num [2] = 000 ... Data to be replaced data [3] @num [3] = 110 data [4] ＾ num [4] = 110 Therefore, the check word that is the lower 3 bits of the sum of these data is check word = 110.

On the other hand, the data after the data exchange is data

[0] = 010 ... num

[0] = 000 ... Data after replacement data [1] = 011 ... num [1] = 001 data [2] = 000 ... num [2] = 010 ... Data after replacement data [3] = 101 ... num [3] = 011 data [4] = 010 ... num [4] = 100

Therefore, data

[0] @num

[0] = 010 ... Data after replacement data [1] @num [1] = 010 data [2] @num [2] = 010 ... Data after replacement data [3] @num [3] = 110 data [4] ＾ num [4] = 110. Therefore, the check word that is the lower 3 bits of the sum of these data is check word = 010.

As described above, for example, when the storage position of the data in the data block is changed due to a malfunction of the device, the check word, that is, the sum of each data changes. It is possible to detect a change in the storage position of the data in the device, and take measures.

The operation of the control means 1 shown in FIG. 1 in this embodiment will be described below with reference to a flowchart. FIG. 11 is a flowchart showing the data initialization operation of the control means 1 in this embodiment. Next, the data initialization operation will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data to create a plurality of data blocks (step S1101).
Next, the control means 1 assigns a different address to each data block (step S1102). Next, the control unit 1 creates a plurality of second data blocks by an exclusive OR of each data block and the corresponding address (step S1103).

Next, the control means 1 writes the data into the storage means 2 (step S1104). Next, the control means 1 creates a check word by summing up the second data blocks (step S1105). Next, the control means 1 writes the check word into the storage means 2 (step S1).
106).

FIG. 12 is a flowchart showing the data check operation of the control means 1 in this embodiment. Next, the data check operation will be described with reference to the flowchart shown in FIG.

First, the control means 1 creates a plurality of data blocks by dividing data in the same manner as in the initialization, and assigns a different address to each data block (step S1201). Next, the control means 1 creates a third data block by exclusive-ORing the address corresponding to the data block (step S1202). Next, the control means 1 creates a fourth data block by summing up the third data block (step S12).
03).

Next, the control means 1 compares the fourth data block with the check word stored in the storage means at the time of initialization (step S1204), and if they match, determines that the data is normal (step S1205). ). Also,
If the comparisons in step S1204 do not match,
It is determined that the data is abnormal (step S1206).

FIG. 13 is a flowchart showing the operation of the data updating process of the control means 1 in this embodiment. Next, the operation of the data update process will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data by the same method as at the time of initialization to create a plurality of data blocks, and assigns different addresses to each data block (step S1301). Next, the control means 1 creates first intermediate data by exclusive-ORing the updated data block with the corresponding address (step S130).
2).

Next, the control means 1 calculates the exclusive OR of the data block before the data update and the corresponding address.
The second intermediate data is created (step S1303).
Next, the control means 1 creates an updated check word according to the equation (20) based on the first and second intermediate data and the address of the check word (step S1304). Next, the control means 1 writes the updated data block and check word into the storage means 2 (step S1305).

Embodiment 4 FIG. 14 is an explanatory diagram showing another embodiment of the data check method according to the present invention. As shown in FIG. 14, the data of the block to be checked is divided into unit lengths (for example, 2 bytes).
Divide into n calculation units. Data of those data
[K] (0 ≦ k ≦ n−1). Also, each calculation unit d
A sequence indicating the relative position of ata [k] in the block is defined as num [k] (0 ≦ k ≦ n−1).

It is assumed that the cycle of the sequence num [k] is equal to or greater than the number n of the divided calculation units. Further, in this embodiment, for example, a sequence of num [k] = k is used as num [k], but another sequence may be used. The check word is stored outside the area of the block to be checked, and is set to checkword here.

In the fourth embodiment, when the following equation is satisfied, it is determined that the data in the block is valid.

[0159]

(Equation 21)

When updating data at an arbitrary position k in a block whose data is guaranteed to be valid by the data check method according to the fourth embodiment, the value checkword 'obtained by the following equation is used. The check word may be updated, and the data may be updated to new data data '[k].

[0161]

(Equation 22)

Next, specific examples will be described below. For example, a case will be described where, in the initialization processing, data is set to “000011010101010”, which is the same as in the first embodiment, and a check word is determined so that each bit becomes an even number in the horizontal parity check.

When this data is divided into three bytes, the following is obtained: 0th data = 000 1st data = 011 1st data = 010 3rd data = 101 4th data = 010

In this case, data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010... Num [4] = 100.

Therefore, in this case, data

[0] + num

[0] = 000 + 000 = 000 data [1] + num [1] = 011 + 001 = 100 data [2] + num [2] = 010 + 010 = 100 data [3] + num [3] = 101 + 011 = 000 data [4] + num [4 ] = 010 + 100 = 110. Therefore, the check word obtained by taking the exclusive OR of these data becomes check word = 110.

In this embodiment, the above-mentioned check word is calculated by the following calculation. Check word = 000/100/100/000/1
10 = 110

In this case, when the data is read, if the check word is 110, the device determines that the data is valid.

Next, the operation when data at an arbitrary position is changed in this data block will be described. For example, the data data [2] of num [2] is changed to “01”.
When changing from “0” to “000”, the changed dat
a ′ [2] = 000.

The data before the change is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010... Num [4] = 100.

Therefore, data

[0] + num

[0] = 000 data [1] + num [1] = 100 data [2] + num [2] = 100... Data to be changed data [3] + num [3] = 000 data [4] + num [4] = 110 Check word = 110

The changed data is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data '[2] = 000... Num [2] = 010... Data after change data [3] = 101. 3] = 011 data [4] = 010... Num [4] = 100.

Therefore, in this case, data

[0] + num

[0] = 000 data [1] + num [1] = 100 data '[2] + num [2] = 010 ... changed data data [3] + num [3] = 000 data [4] + num [4] = 110. Therefore, the check word obtained by taking the exclusive OR of these data is check word = 000.

In this embodiment, the check word after the data change is calculated by the following operation. Check word after change = 110 / (010 + 010) / (000 + 010) = 110/100/010 = 000

As described above, when data is changed, a new check word is not obtained from all data,
An extremely simple calculation can be performed based on the data to be changed, and the calculation can be speeded up.

Next, the operation when data exchange is performed will be described. For example, when the 0th data and the 2nd data are exchanged, the data before the data exchange is data

[0] = 000… num

[0] = 000 data [1] = 011... Num [1] = 001 data [2] = 010... Num [2] = 010 data [3] = 101... Num [3] = 011 data [4 ] = 010 ... num [4] = 100,

[0176] Met.

On the other hand, the data after data exchange is data

[0] = 010 ... num

[0] = 000 data [1] = 011 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 101 ... num [3] = 011 data [4 ] = 010... Num [4] = 100.

Therefore, data

[0] + num

[0] = 010 data [1] + num [1] = 100 data [2] + num [2] = 010 data [3] + num [3] = 000 data [4] + num [4] = 110. The check word obtained by taking the exclusive OR of these data is check word = 010.

As described above, for example, when the storage position of the data in the data block is changed due to a malfunction of the device, the check word, that is, the value obtained by taking the exclusive OR of each data changes. By examining, it is possible to detect a change in the storage position of the data in the data block, and take measures.

The operation of the control means 1 shown in FIG. 1 in this embodiment will be described below with reference to a flowchart. FIG. 15 is a flowchart showing the data initialization operation of the control means 1 in this embodiment. Next, the data initialization operation will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data to create a plurality of data blocks (step S1501).
Next, the control means 1 assigns a different address to each data block (step S1502). Next, the control means 1 creates a plurality of second data blocks by adding the addresses corresponding to the respective data blocks (step S1503).

Next, the control means 1 writes the data into the storage means 2 (step S1504). Next, the control means 1 creates a check word by taking the exclusive OR of all the second data blocks (step S1505). Next, the control unit 1 writes the check word into the storage unit 2 (Step S1506).

FIG. 16 is a flowchart showing a data check operation of the control means 1 in this embodiment. Next, the data check operation will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data by the same method as at the time of initialization, creates a plurality of data blocks, and assigns different addresses to each data block (step S1601). Next, the control unit 1 creates a third data block by adding an address corresponding to the data block (step S1602). Next, the control means 1 creates a fourth data block by taking the exclusive OR of all the third data blocks (step S1603).

Next, the control means 1 stores the memory means 2 at the time of initialization.
Is read from the storage means 2 and the check word is compared with the value of the fourth data block (step S1604). If they match, it is determined that the data is normal (step S1605). Also,
If the comparisons in step S1604 do not match,
It is determined that the data is abnormal (step S1606).

FIG. 17 is a flowchart showing the operation of the data updating process of the control means 1 in this embodiment. Next, the operation of the data update process will be described with reference to the flowchart shown in FIG.

First, the control means 1 divides data by the same method as at the time of initialization to create a plurality of data blocks, and assigns different addresses to each data block (step S1701). Next, the control means 1 creates the first intermediate data by adding the corresponding address to the updated data block (step S170).
2). Next, the control unit 1 creates the second intermediate data by adding the corresponding address to the data block before the data update (step S1703).

Next, the control means 1 creates an updated check word according to the formula (22) based on the first and second intermediate data and the check word before update (step S17).
04). Next, the control means 1 writes the updated data block and check word into the storage means 2 (step S1).
705).

Embodiment 5 FIG. In the first embodiment, when the entire area becomes 0 due to a failure of the memory device or the like, the entire area becomes 0 unless the following equation is satisfied.
Can be detected.

[0190]

(Equation 23)

For example, in order that this equation does not hold, a
When = 0, n may be 4 or more.

Next, specific examples will be described below. For example, in the initialization process, all data is set to “00000000”.
00000000 "will be described.
Also, let a = 000.

When this data is divided into three bytes, the following is obtained: 0th data = 000 1st data = 000 2nd data = 000 3rd data = 000 4th data = 000

In this case, if the position of the check word is j = 5, for example, data

[0] = 000… num

[0] = 000 data [1] = 000 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 000 ... num [3] = 011 data [4 ] = 000... Num [4] = 100.

Therefore, in this case, Becomes

Accordingly, the check word is data [5] = 000 + {inverted (010 + 000) +1} 101 = 000 + {101 + 1} 101 = 000 + 110} 101 = 011.

Therefore, data [5] ＾ num [5] = 011 ＾ 101 = 11
It becomes 0.

Therefore, in this case, data

[0] @num

[0] = 000 data [1] @num [1] = 001 data [2] @num [2] = 010 data [3] @num [3] = 011 data [4] @num [4] = 100 data [5] $ num [5] = 110 ... Check words Total = 000.

When the above data is read, a = 00
If it is 0, the device determines that this data is valid. Therefore, in this case, the total value indicates a, and a = 00
Since it is 0, the data is determined to be valid.

On the other hand, due to the failure of the memory device, etc.
When all data including the check word becomes 0, data

[0] = 000… num

[0] = 000 data [1] = 000 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 000 ... num [3] = 011 data [4 ] = 000... Num [4] = 100 data [5] = 000... Num [5] = 101...

Therefore, in this case, data

[0] @num

[0] = 000 data [1] @num [1] = 001 data [2] @num [2] = 010 data [3] @num [3] = 011 data [4] @num [4] = 100 data [5] ＾ num [5] = 101... Check word total = 111.

When the above data is read, a = 00
If it is 0, the device determines that this data is valid. If a is not equal to 000, it is determined that the data is invalid data generated due to a failure of the memory device or the like. In this case, since the total value indicates a, a = 111, and a = 000 is not satisfied, it is determined that the data is invalid.

As described above, it is possible to detect a situation in which all data becomes 0 due to a failure of a memory device or the like, so that it is possible to take measures.

Embodiment 6 FIG. In the second embodiment, when the entire area becomes 0 due to a memory device failure or the like, the entire area becomes 0 unless the following equation is satisfied.
Can be detected.

[0205]

(Equation 24)

For example, in order that this equation does not hold, a
In the case of = 0, n may be other than a multiple of 4.

Next, specific examples will be described below. For example, in the initialization process, all data is set to “00000000”.
00000000 "will be described.
Also, let a = 010.

When this data is divided into three bytes, the following is obtained: 0th data = 000 1st data = 000 2nd data = 000 3rd data = 000 4th data = 000

In this case, if the position of the check word is j = 5, for example, data

[0] = 000… num

[0] = 000 data [1] = 000 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 000 ... num [3] = 011 data [4 ] = 000... Num [4] = 100.

Accordingly, in this case, Becomes

Therefore, the check word is data [5] = {010/000/001/010/011/100} + {inversion 101 + 1} = 010/100 + (010 + 1) = 110 + 011 = 001.

Therefore, the check word is data [5] + num [5] = 001 + 101 = 11
It becomes 0.

Accordingly, in this case, data

[0] + num

[0] = 000 data [1] + num [1] = 001 data [2] + num [2] = 010 data [3] + num [3] = 011 data [4] + num [4] = 100 data [5] + num [5] = 110... Check word EXOR = 010.

When the above data is read, a = 00
If it is 0, the device determines that this data is valid. Therefore, in this case, all the values of the exclusive OR are a
, And since a = 010, the data is determined to be valid.

On the other hand, due to the failure of the memory device, etc.
When all data including the check word becomes 0, data

[0] = 000… num

[0] = 000 data [1] = 000 ... num [1] = 001 data [2] = 000 ... num [2] = 010 data [3] = 000 ... num [3] = 011 data [4 ] = 000... Num [4] = 100 data [5] = 000... Num [5] = 101...

Therefore, in this case, data

[0] + num

[0] = 000 data [1] + num [1] = 001 data [2] + num [2] = 010 data [3] + num [3] = 011 data [4] + num [4] = 100 data [5] + num [5] = 101... Check word EXOR = 001.

When the above data is read, a = 01
If it is 0, the device determines that this data is valid. If a is not equal to 010, it is determined that the data is invalid data generated due to a memory device failure or the like. In this case, the values of all exclusive ORs indicate a,
Since a = 001 and a = 010 is not satisfied, it is determined that the data is invalid.

As described above, it is possible to detect a situation where all data becomes 0 due to a failure of the memory device or the like, so that it is possible to take measures.

[0219]

As described above, according to the first aspect of the present invention, by examining the value of the checksum, it is possible to detect a change in the data storage position in the data block, and to take measures. This has the effect.

According to the second aspect of the present invention, a new check word is not obtained from all data but is obtained by an extremely simple operation based on data to be changed, so that the operation can be speeded up. This has the effect.

According to the third aspect of the present invention, by examining the value of the horizontal parity, it is possible to detect a change in the storage position of the data in the data block, and to take measures.

According to the fourth aspect, when data is changed, a new check word is not obtained from all data but is obtained by an extremely simple calculation based on the data to be changed. This has the effect of realizing the conversion.

Further, according to the fifth aspect, by examining the check word, it is possible to detect a change in the storage position of the data in the data block, and to take measures.

According to the sixth aspect of the present invention, when data is changed, a new check word is not obtained from all data but is obtained by an extremely simple operation based on data to be changed. This has the effect of realizing the conversion.

Further, according to the seventh aspect, by checking the check word, it is possible to detect a change in the storage position of the data in the data block, and to take an effect that a countermeasure can be taken.

According to the eighth aspect, when data is changed, a new check word is not obtained from all data but is obtained by an extremely simple calculation based on the data to be changed. This has the effect of realizing the conversion.

Further, according to the ninth aspect, by checking the value of the checksum, it is possible to detect a situation in which the data becomes all 0 due to a failure of the memory device or the like. Has the effect of being possible

Further, according to the tenth aspect, by checking the value of the horizontal parity, it is possible to detect a situation where all data becomes 0 due to a failure of a memory device or the like. It has the effect that it becomes possible.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a configuration for performing data check according to the related art and the present invention.

FIG. 2 is an explanatory diagram showing one embodiment of a data check method according to the present invention.

FIG. 3 is a flowchart showing an operation of data initialization of the control means 1 according to the first embodiment.

FIG. 4 is a flowchart showing a data check operation of the control means 1 according to the first embodiment.

FIG. 5 is a flowchart showing an operation of a data updating process of the control means 1 according to the first embodiment.

FIG. 6 is an explanatory diagram showing another embodiment of the data check method according to the present invention.

FIG. 7 is a flowchart showing an operation of data initialization of the control means 1 according to the second embodiment.

FIG. 8 is a flowchart showing a data check operation of the control means 1 according to the second embodiment.

FIG. 9 is a flowchart showing an operation of a data updating process of the control means 1 according to the second embodiment.

FIG. 10 is an explanatory diagram showing another embodiment of the data check method according to the present invention.

FIG. 11 is a flowchart showing an operation of data initialization of the control means 1 according to the third embodiment.

FIG. 12 is a flowchart showing a data check operation of the control means 1 according to the third embodiment.

FIG. 13 is a flowchart showing an operation of a data update process of the control means 1 according to the third embodiment.

FIG. 14 is an explanatory diagram showing another embodiment of the data check method according to the present invention.

FIG. 15 is a flowchart showing an operation of data initialization of the control means 1 according to the fourth embodiment.

FIG. 16 is a flowchart showing a data check operation of the control means 1 according to the fourth embodiment.

FIG. 17 is a flowchart showing an operation of a data updating process of the control means 1 according to the fourth embodiment.

FIG. 18 is an explanatory diagram showing an example of a conventional data check method using a checksum.

FIG. 19 is an explanatory diagram showing another example of a conventional data check method using a checksum.

FIG. 20 is an explanatory diagram showing an example of a conventional data check method using a horizontal parity check.

FIG. 21 is an explanatory diagram showing another example of a conventional data check method using a horizontal parity check.

[Explanation of symbols]

 1 control means 2 storage means

Claims (10)

[Claims] 1. A data dividing step of dividing data by a predetermined length to create a plurality of first data blocks, and an address allocating step of allocating an address to each of the plurality of first data blocks. Creating a plurality of second data blocks by performing an exclusive OR operation on a value of the first data block and a corresponding address for each of the plurality of first data blocks Process and data reliability checks
A first check word insertion step of inserting or adding the check word at a predetermined address within the address range of the first data block, and a second sum obtained by summing the plurality of second data blocks. Take the two's complement of the sum of the data blocks, then take the exclusive OR of the two's complement of this sum with the address of the first check word, and further take the value resulting from the exclusive OR. A first check word creating step of creating the first check word by adding a predetermined value; a writing step of writing the first data block and the first check word into storage means; When the plurality of first data blocks and the first check word are read from the storage means, the first data block is read from each of the plurality of first data blocks. A third data block creation step of taking an exclusive OR of the value of the block and the corresponding address to create a plurality of third data blocks; and performing an exclusive OR of the address of the first check word and the corresponding address. A second check word creating step of taking a logical sum to create a second check word; and taking a sum of all of the plurality of third data blocks and the second check word to form a fourth check word.
A data check method, comprising: a fourth data block creation step of creating a data block of (i), and a check step of determining whether or not the fourth data block is the same as the predetermined value. 2. A first intermediate data creation step of taking an exclusive OR of a value of a first check word before update and a corresponding address when updating data, and creating first intermediate data. And a first data block to be changed among the plurality of first data blocks.
XORing the value of the data block with the corresponding address to create a plurality of second intermediate data, a second intermediate data creating step, and the changed first data block and the first check A third intermediate data generating step of obtaining an exclusive OR with a word address to generate third intermediate data; and a step of generating the first intermediate data, the plurality of second intermediate data, and the third intermediate data. And taking the exclusive OR of the sum and the address of the first check word to create a first check word after updating the data. 2. The data check method according to claim 1, wherein: 3. A data dividing step of dividing data by a predetermined length to generate a plurality of first data blocks, and an address allocating step of allocating an address to each of the plurality of first data blocks. A second data block creating step of adding a corresponding address to the value of the first data block for each of the plurality of first data blocks to create a plurality of second data blocks; Inserting a first check word into a predetermined address within an address range of the first data block to add or check a reliability of the first data block; and inserting a first check word into the plurality of second data blocks. A third data block creation step of taking all exclusive ORs to create a third data block; A fourth data block creation step of taking an exclusive OR with a determined value to create a fourth data block; and taking a 2's complement of the address of the first check word to obtain a fourth data block. A writing step of writing the first data block and the first check word into storage means; and a writing step of writing the first data block and the first check word into storage means. When reading the block and the first check word from the storage means, for each of the plurality of first data blocks, add a corresponding address to the value of the first data block, and A fifth data block creating step of creating a fifth data block, and adding a corresponding address to the value of the first check word to create a second check word A second check word creating step of: performing an exclusive OR operation of all of the plurality of fifth data blocks with the second check word to create a sixth data block; A check step of determining whether or not the sixth data block is equal to the predetermined value. 4. When data is updated, a first intermediate data creating step of adding a corresponding address to a value of a first check word before updating to create first intermediate data, A second intermediate block for creating second intermediate data by adding a corresponding address to a value of the first data block to be changed in the first data block, the value of the first data block. A data creating step, a third intermediate data creating step of adding a corresponding address to the changed value of the first data block to create third intermediate data, and the first intermediate data and the second intermediate data A fourth intermediate data creating step of creating an exclusive OR of the intermediate data of the third intermediate data and the third intermediate data to create fourth intermediate data;
4. A data check method according to claim 3, further comprising the step of: adding a two's complement of the address of said first check word to said fourth intermediate data to create a check word after data update. . 5. A data dividing step of dividing data by a predetermined length to generate a plurality of first data blocks, and an address allocating step of allocating an address to each of the plurality of first data blocks. Creating a plurality of second data blocks by performing an exclusive OR operation on a value of the first data block and a corresponding address for each of the plurality of first data blocks And a step of dividing the plurality of first data blocks into a first data block.
A first data block storing step of storing in a storage means of
A check word creating step of calculating a sum of the plurality of second data blocks and creating a check word based on the obtained sum, a check word storing step of storing the check word in second storage means, Multiple first
When reading out the data block from the first storage means, the first data block
A third data block creating step of taking an exclusive OR of the value of the data block and the corresponding address to create a plurality of third data blocks; and taking a total sum of the plurality of third data blocks to obtain a fourth data block. And a check step of determining whether or not the fourth data block is the same as the check word read from the second storage means. Data check method. 6. When data is updated, a first intermediate data creating step of taking an exclusive OR of the changed first data block and a corresponding address to create first intermediate data; An exclusive OR of a value of the first data block and an address corresponding to the first data block before the change to be changed among the plurality of first data blocks is calculated. A second intermediate data generating step of generating intermediate data; a sum of the first intermediate data, a two's complement of the second intermediate data, and a value of a check word before updating; 6. The data check method according to claim 5, further comprising the step of: 7. A data dividing step of dividing data by a predetermined length to create a plurality of first data blocks, and an address assigning step of assigning an address to each of the plurality of first data blocks. A second data block creating step of adding a corresponding address to a value of the first data block for each of the plurality of first data blocks to create a plurality of second data blocks; A first data block storing step of storing a plurality of first data blocks in a first storage means, and an exclusive OR of the plurality of second data blocks is calculated based on a result of the exclusive OR. A check word creating step of creating a check word, and a check word storing step of storing the check word in the second storage means.
When reading out the plurality of first data blocks from the first storage means, a value corresponding to the value of the first data block is added to each of the first data blocks to obtain a plurality of first data blocks. A third data block creating step of creating a third data block, and a fourth data block creating step of taking an exclusive OR of all of the plurality of third data blocks to create a fourth data block. A check step of determining whether or not the fourth data block is the same as a check word read from the second storage means. 8. When data is updated, a plurality of first
A first intermediate block for generating a plurality of first intermediate data by adding a corresponding address to a value of the first data block to be changed among the first data blocks of the first data block A data creation step, a second intermediate data creation step of adding a corresponding address to the changed value of the first data block to create second intermediate data,
The exclusive OR of the value of the check word before update, the first intermediate data, and the second intermediate data is calculated,
Creating a check word after updating the data. 9. The checking step includes: when reading out the first data block and the first check word from the storage means, if all the values of the first data block are all 0, The exclusive OR of the plurality of third data blocks created by taking the exclusive OR of the value of the data block and the corresponding address, and the exclusive OR of the address corresponding to the first check word 2. The data check method according to claim 1, wherein it is determined whether the sum of the created second check word and the second check word is the same as a predetermined value. 10. The checking step includes: when reading out the first data block and the first check word from the storage means, when all the values of the first data block are all 0, A second check created by adding all of the plurality of third data blocks created by adding the corresponding addresses to the data block values and the addresses corresponding to the first check words, respectively. 4. The data check method according to claim 3, wherein it is determined whether or not the exclusive OR of the word and the word is equal to a predetermined value.