WO2017088507A1

WO2017088507A1 - Method and device for updating error checking and correction code

Info

Publication number: WO2017088507A1
Application number: PCT/CN2016/091840
Authority: WO
Inventors: 黄生儒; 廉肖洁; 薛军
Original assignee: 华为技术有限公司
Priority date: 2015-11-26
Filing date: 2016-07-27
Publication date: 2017-06-01
Also published as: CN106802837B; CN106802837A

Abstract

Disclosed in an embodiment of the invention is a method for updating an error checking and correction (ECC) code. The method is applied in a process of writing data to a storage medium comprising multiple storage blocks. The method comprises: acquiring an ECC code for a total amount of data in each storage block; when a byte of data is randomly written to the storage medium, acquiring the written data, a location in a storage space where the written data is stored, and initial data stored in the storage space prior to the writing of the data; performing a logic operation to acquire a column check code of an ECC code for the total amount of data in the storage block to which the storage space that stored the written byte of data belongs, and revising, according to the written data and the initial data, a row check code of the ECC code for the total amount of data in the storage block to which the storage space belongs; and obtaining an updated ECC code according to the column check code of the ECC code obtained via the operation and the revised row check code of the ECC code. The invention can protect random read and write performance of a storage medium from being degraded when an ECC check is performed.

Description

Method and device for updating error detection and correcting ECC code

This application claims priority to Chinese Patent Application No. 201510844082.2, entitled "A Method and Apparatus for Updating Error Detection and Correcting ECC Codes", filed on November 26, 2015, the entire contents of which are hereby incorporated by reference. The citations are incorporated herein by reference.

Technical field

The present invention relates to the field of computer technologies, and in particular, to a method and apparatus for updating an error detection and correcting an ECC code.

Background technique

In the computer field, the two mainstream flash memories are NAND Flash (internal memory unit with NAND gate connected flash memory) and NOR Flash (internal memory unit with NAND gate connected flash memory). Among them, NAND Flash is prone to bit inversion problem. If bit error is reversed in important program code or key data in NADN Flash, the computer system may not operate normally. Therefore, in order to improve reliability, NAND Flash application is required. Perform ECC (Error Checking and Correction) check.

ECC check refers to calculating the ECC code by means of a data block (such as 256 bytes or 512 bytes of data as one data block) when reading and writing data, by comparing the ECC code of the written data with the read data. Whether the ECC codes are consistent or not, and whether the data is bit-reversed is determined. Since NAND Flash itself reads and writes data in the form of data blocks, ECC verification is perfectly suitable for NAND Flash.

Further, as the size of the memory process continues to shrink, NOR Flash is also more and more prone to bit inversion problems. Currently, ECC verification is introduced to NOR Flash. However, unlike NAND Flash reading and writing, NOR Flash is a random read/write method, that is, the size of read and write data is random. If ECC is directly introduced into NOR Flash, it will be limited to reading in blocks. Writing data destroys the random read and write of NOR Flash.

Summary of the invention

Embodiments of the present invention provide a method and apparatus for updating an error detection and correcting an ECC code, which can achieve random read and write performance of a storage medium when performing ECC verification.

A first aspect of the embodiments of the present invention provides a method for updating an ECC code, where the method is applied to a data writing process of a storage medium, where the storage medium includes a plurality of storage blocks, and the method includes:

Obtaining an ECC code of total data in each of the storage blocks, where the ECC code includes a column check code and a row check code;

When a byte of data is randomly written to the storage medium, the written data, the location of the storage space where the data is written, and the original data stored by the storage space before the data is written are acquired. Wherein the location of the storage space is used to determine a storage block to which the storage space belongs;

And performing, by using a column check code of the ECC code of the written data, a column check code of the ECC code of the original data, and a column check code of an ECC code of the total data in the storage block to which the storage space belongs. a logical operation of obtaining a column check code of an ECC code of total data in a storage block to which the storage space belongs after writing one-byte data, and modifying the data according to the written data and the original data a row check code of the ECC code of the total data in the storage block to which the storage space belongs;

a column check code of the ECC code of the total data in the memory block to which the storage space belongs after the one-byte data is written, and an ECC code of the total data in the memory block to which the modified storage space belongs The check code is obtained, and the ECC code of the total data in the storage block to which the updated storage space belongs is obtained.

In a first possible implementation manner of the first aspect, the column check code of the ECC code of the written data, the column check code of the ECC code of the original data, and the storage space belong to The column check code of the ECC code of the total data in the storage block is logically operated, and the column check code of the ECC code of the total data in the storage block to which the storage space belongs after the one-byte data is written is obtained, including:

Calculating a column check code of the ECC code of the written data, and a column check code of the ECC code of the original data; according to a formula

Calculating a column check code of an ECC code of total data in a storage block to which the storage space belongs after writing one-byte data, wherein

a column check code indicating an ECC code of the calculated total data, the ECC_CP indicating a column check code of an ECC code of the total data before calculation, and the Data_ECC_CP indicating an ECC code of the written data A column check code, the Buf_ECC_CP indicating a column check code of the ECC code of the original data. Among them, the formula

The principle is that since ^ is both an increase and a removal in the XOR algorithm, the formula is equivalent to adding Data_ECC_CP to ECC_CP and removing Buf_ECC_CP.

In conjunction with the first possible implementation of the first aspect, in a second possible implementation, the root And modifying the row check code of the ECC code of the total data in the storage block to which the storage space belongs according to the data to be written and the original data, including:

Determining a parity of the written data and a parity of the original data; determining whether a parity of the written data is the same as a parity of the original data; if yes, modifying the total data a row check code of the ECC code; if not, determining a bit associated with the data in the written storage space in the row check code of the ECC code of the total data, and reversing the bit Turning, wherein the inversion refers to changing data 0 to data 1, or changing data 1 to data 0.

In conjunction with the first aspect and the first or second possible implementation of the first aspect, in a third possible implementation, the memory block is 256 bytes of storage space. On the one hand, the smaller the number of bytes of the storage block, the more the number of storage blocks in the storage medium, the more ECC codes to be recorded. On the other hand, the more bytes of the storage block, the total in the storage block. The larger the data, the more complex the ECC code to be recorded, and the combination of the above two factors, the most appropriate setting of 256 bytes.

A second aspect of the embodiments of the present invention provides a method for verifying an ECC code, where the method is applied to a data writing process of a storage medium, where the storage medium includes a plurality of storage blocks, and the method includes:

When randomly reading data to the storage medium, performing parity check on the read data; determining whether the parity is in error; if yes, totaling in a storage block to which the read data is located The data is subjected to ECC code verification.

In a first possible implementation manner of the second aspect, the performing parity check on the data in the storage space where the read data is located includes:

Calculating a parity code of the read data; querying a parity code stored in a storage space of the read data when writing data; determining whether the calculated parity code is related to the saved parity The parity is the same; if not, the parity is erroneous; if so, the parity is not erroneous. With this technical solution, it is possible to first determine whether the read data is in error.

In a second possible implementation manner of the second aspect, the performing the ECC code check on the total data in the storage block to which the storage space of the read data belongs includes:

Calculating an ECC code of the total data in the storage block to which the read data belongs, and querying a parity code saved when the storage block to which the read data belongs is written in the data; The ECC code is compared with the saved ECC code to determine the reading The bit-inverted bit occurs in the data, and is inverted, wherein the inversion refers to changing data 0 to data 1, or changing data 1 to data 0. With this technical solution, it is possible to determine the bit of the error in the read data and correct it.

A third aspect of the embodiments of the present invention provides an apparatus for updating an ECC code, where the apparatus has a behavior function for implementing the method provided by the foregoing first aspect, and the function may be implemented by using hardware or by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the functions described above.

A fourth aspect of the embodiments of the present invention provides an apparatus for verifying an ECC code, where the apparatus has a behavior function for implementing the method provided by the foregoing second aspect, and the function may be implemented by hardware, or may be implemented by hardware. . The hardware or software includes one or more modules corresponding to the functions described above.

A fifth aspect of the embodiments of the present invention provides a storage device including a processor, a communication bus, a storage medium, a memory, and an input/output interface. The communication bus is configured to implement connection communication between the components; the storage medium includes a plurality of virtual storage blocks for storing data written by the external device, or sending data read by the external data; the input and output The interface is used to provide an interface to an external device for reading and writing data; the program code is stored in the memory.

In one aspect, when the storage device is in the data writing process, the processor is configured to call the program code stored in the memory, and performs the following operations:

Acquiring an ECC code of total data in each of the storage blocks, the ECC code including a column check code and a row check code; when randomly writing one byte of data to the storage medium, acquiring the written data The location of the storage space in which the data is written and the storage space is stored before the data is written Raw data, wherein the location of the storage space is used to determine a storage block to which the storage space belongs; a column check code of an ECC code for the written data, and a column of an ECC code of the original data The check code and the column check code of the ECC code of the total data in the storage block to which the storage space belongs are logically operated, and the ECC of the total data in the storage block to which the storage space belongs after the one-byte data is written is obtained. a column check code of the code, and modifying a row check code of the ECC code of the total data in the memory block to which the storage space belongs according to the written data and the original data; writing a byte according to the After the data, the column check code of the ECC code of the total data in the storage block to which the storage space belongs and the row check code of the ECC code of the total data in the storage block to which the modified storage space belongs are updated. The ECC code of the total data in the storage block to which the storage space belongs.

Optionally, the processor passes the column check code of the ECC code of the written data, the column check code of the ECC code of the original data, and the ECC code of the total data in the storage block to which the storage space belongs. The column check code performs a logical operation, and the specific operation of the column check code of the ECC code of the total data in the storage block to which the storage space belongs after the one-byte data is written is: calculating the written data Column check code of the ECC code, and column check code of the ECC code of the original data; according to the formula

a column check code indicating an ECC code of the calculated total data, the ECC_CP indicating a column check code of an ECC code of the total data before calculation, and the Data_ECC_CP indicating an ECC code of the written data Column check code, the Buf_ECC_CP represents a column check code of the ECC code of the original data, and the ^ represents a bitwise exclusive OR operation.

Optionally, the specific operation of the processor to modify the row check code of the ECC code of the total data in the storage block to which the storage space belongs according to the written data and the original data is: determining the write The parity of the data, and the parity of the original data; determining whether the parity of the written data is the same as the parity of the original data; if so, the row of the ECC code of the total data is not modified a check code; if not, determining a bit associated with the data in the written storage space in the row check code of the ECC code of the total data, and inverting it, wherein The inversion refers to changing data 0 to data 1, or changing data 1 to data 0.

On the other hand, when the storage device is in the data reading process, the processor is used to call the program code stored in the memory, and performs the following operations:

When randomly reading data to the storage medium through the input/output interface, performing parity check on the read data; determining whether the parity is in error; if yes, belonging to a storage space where the read data is located The total data in the memory block is subjected to ECC code verification.

Optionally, the specific operation of the processor to perform parity check on the data of the storage space where the read data is located is: calculating a parity code of the read data; querying a storage space where the read data is written a parity code saved when data is entered; determining whether the calculated parity code is the same as the saved parity code; if not, the parity error occurs; if yes, the parity is not errored .

Optionally, the specific operation of performing ECC code verification on the total data in the storage block to which the read data belongs is: calculating a total of the storage blocks to which the read data belongs. An ECC code of the data; querying a parity code saved when the storage block to which the read data belongs is written in the data; determining the calculated ECC code by comparing the calculated ECC code with the saved ECC code A bit inverted bit occurs in the read data, and is inverted, wherein the inversion refers to changing data 0 to data 1, or changing data 1 to data 0.

It can be seen that, in the embodiment of the present invention, when the data is randomly written to the storage medium, the ECC code of the total data in the storage block can be modified only according to the written data and the original data before the writing, which can be solved in the prior art. The data of one storage block must be completely written before the problem of the ECC code can be calculated, so that the random writeability of the storage medium is not damaged. In the embodiment of the present invention, when the data is randomly read from the storage medium, the read data is first performed. Parity check, only when the parity error occurs, the ECC check is performed, which can solve the problem that the ECC check can be performed after the data of one memory block is completely read every time the data is read in the prior art, thereby reducing the problem. Destruction of random readability of storage media.

DRAWINGS

BRIEF DESCRIPTION OF THE DRAWINGS In the following, the embodiments of the present invention will be briefly described, and the drawings in the following description are merely illustrative of some embodiments of the present invention. Other drawings may also be obtained from those of ordinary skill in the art in light of the inventive work.

1 is a schematic flowchart of a method for updating an ECC code according to an embodiment of the present invention;

2 is a schematic flowchart of a method for verifying an ECC code according to an embodiment of the present invention;

3 is a schematic structural diagram of an apparatus for updating an ECC code according to an embodiment of the present invention;

4 is a schematic structural diagram of a row and column code modification module according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of an apparatus for verifying an ECC code according to an embodiment of the present invention;

6 is a schematic structural diagram of a parity check module according to an embodiment of the present invention;

7 is a schematic structural diagram of an ECC code verification module according to an embodiment of the present invention;

FIG. 8 is a schematic structural diagram of a storage device according to an embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

In order to facilitate the understanding of the embodiments of the present invention, the calculation method of the ECC code will be described first. The ECC code includes a column check code and a row check code, and the calculation object is a data block. The following is an example of 256 bytes of data as a data block:

First, calculate the column check code of the ECC code.

It should be understood that 1 byte (Byte) includes 8 bits (Bit), then 256 bytes are expanded in a bitwise manner to obtain Table 1.

Table 1

CP0 represents the column polarity of the 0th column, the 2nd column, the 4th column, and the 6th column;

CP1 indicates the column polarity of the first column, the third column, the fifth column, and the seventh column;

CP2 indicates the column polarity of the 0th column, the 1st column, the 4th column, and the 5th column;

CP3 indicates the column polarity of the second column, the third column, the sixth column, and the seventh column;

CP4 indicates the column polarity of the 0th column, the 1st column, the 2nd column, and the 3rd column;

CP5 indicates the column polarities of the fourth column, the fifth column, the sixth column, and the seventh column.

Where CP0=[Bit0]^[Bit2]^[Bit4]^[Bit6], ^ represents a bitwise XOR operation, [Bit0] represents the result of 256 Bit0 XORs of column 0, [Bit2] indicates the first The result of 256 Bit2 XOR of 2 columns, [Bit4] represents the result of 256 Bit4 XOR of the 4th column, and [Bit6] represents the result of 256 Bit6 XOR of the 6th column.

By analogy, CP1 to CP5 can be obtained. Assuming that ECC_CP indicates the column check code of the ECC code, ECC_CP = {CP5, CP4, CP3, CP2, CP1, CP0}.

Second, calculate the row check code of the ECC code.

Similarly, 256 bytes are expanded in a bitwise manner to obtain Table 2.

Table 2

RP0 is the row polarity of rows 0, 2, 4, 6, ... 252, 254;

RP1 is the row polarity of the first, third, fifth, seventh, ..., 253, and 255 lines;

RP2 is the row polarity of lines 0, 1, 4, 5, ... 252, 254 (processing two lines, skipping two lines);

RP3 is the row polarity of the 2nd, 3rd, 6th, 7th, ..., 254th, and 255th lines (skip two lines, processing two lines);

RP4 is the row polarity of the line (processing four lines, skipping four lines);

RP5 is the row polarity of the line (skip four lines, handle four lines);

RP6 is the row polarity of the line (processing eight lines, skipping eight lines);

RP7 is the row polarity of the line (skip eight lines, handle eight lines);

RP8 is the row polarity of the line (processing sixteen lines, skipping sixteen lines);

RP9 is the row polarity of the line (skip six lines, handle sixteen lines);

RP10 is the row polarity of the line (processing thirty-two lines, skipping thirty-two lines);

RP11 is the row polarity of the line (skip thirty-two lines, processing thirty-two lines);

RP12 is the row polarity of the line (processing sixty-four lines, skipping sixty-four lines);

RP13 is the row polarity of the line (skip sixty-four lines, processing sixty-four lines);

RP14 is the row polarity of the line (handling one hundred and twenty-eight lines, skipping one hundred and twenty-eight lines);

RP15 is the line polarity of the line (skip one hundred and twenty-eight lines, handle one hundred and twenty-eight lines);

Where RP0=[Byte0]^[Byte2]^[Byte4]^[Byte6]^......^[Byte252]^[Byte254], ^ represents the operation of XOR, [Byte0] represents the 0th line The result of 8 Bit XOR, [Byte2] represents the result of the 8 Bit XOR of the 2nd line, and so on.

By analogy, RP1 ~ RP15 can be obtained. Assuming that ECC_RP indicates the row check code of the ECC code, ECC_RP = {RP15, RP14, ..., RP3, RP2, RP1, RP0}.

In summary, the ECC code includes a column check code ECC_CP and a row check code ECC_RP. In the ECC check process of the storage medium, when writing data, it is necessary to write data in units of 256-byte data blocks, that is, each When an 256-byte data is written, an ECC code is calculated. Similarly, when reading data, it is necessary to read data in units of 256-byte data blocks, that is, an ECC code is calculated every time 256-byte data is read. However, a random read/write storage medium such as NOR Flash performs random read and write in units of 1 byte when reading and writing data. One of the advantages is that small data of 256 bytes or less can be read and written at will. If an ECC check is directly introduced into a random read/write storage medium, the advantage will no longer exist.

1 is a schematic flowchart of a method for updating an ECC code according to an embodiment of the present invention. The method is applied to a data writing process of a storage medium, where the storage medium includes a plurality of storage blocks, optionally, in the embodiment of the present invention. Set the storage block to 256 bytes of storage space. The flow of the method for updating the ECC code in this embodiment as shown in the figure may include:

S101. Obtain an ECC code of total data in each storage block.

Specifically, the ECC code corresponding to the 256-byte data in each memory block can be calculated by the calculation method of the ECC code introduced above. Optionally, if the ECC code calculated after the last data write is recorded locally, or the newly calculated ECC code in other cases, the ECC code is directly found, and does not need to be calculated again. It should be noted that the purpose of obtaining the ECC code of the total data in each memory block is to obtain a new ECC code by modifying the ECC code when the data is randomly written to the storage medium.

S102, when randomly writing one byte of data to the storage medium, acquiring the written data, the location of the storage space where the data is written, and the storage space before the writing of the data by the storage space Raw data.

It should be pointed out that only the case of writing one byte of data is introduced here, but it should be understood that in the case of data writing, generally, data of multiple bytes is written. For this case, only one The byte mode can be executed multiple times.

Specifically, when one byte of data is randomly written to the storage medium, the written data, the location of the storage space where the data is written, and the original data stored by the storage space before the data is written are acquired. The location of the storage space is used to determine a storage block to which the storage space belongs. Optionally, the location of the storage space may be a storage address or a serial number of a byte. Take the serial number of the storage space as an example: assume that the data randomly written to the storage medium is 0xF0, and the byte corresponding to the storage space to which the data is written is Byte9, and the storage space is before the data is written. If the stored original data is 0x31, the obtained information is 0xF0, Byte9, and 0x31. Further, it is assumed that Byte 0 to Byte255 belong to the first storage block, and since Byte9 falls in the interval of Byte 0 to Byte 255, it can be determined that the data is The written storage space belongs to the first storage block.

S103, a column check code of an ECC code of the written data, a column check code of an ECC code of the original data, and a column check of an ECC code of a total data in a storage block to which the storage space belongs The code performs a logical operation to obtain a column check code of the ECC code of the total data in the storage block to which the storage space belongs after writing the one-byte data, and is modified according to the written data and the original data. A row check code of an ECC code of total data in a storage block to which the storage space belongs.

Specifically, it can be seen from the above that the ECC code includes a column check code and a row check code, so it is necessary to separately modify the ECC code of the total data in the storage block to which the written byte belongs according to the written data and the original data. The row check code of the column check code and the ECC code.

First, the method of modifying the column check code of the ECC code may be: a column check code of the ECC code of the written data, a column check code of the ECC code of the original data, and a total data in the storage block to which the storage space belongs. The column check code of the ECC code performs a logical operation to obtain a column check code of the ECC code of the total data in the memory block to which the storage space to which the one-byte data is written. The specific implementation process is:

Step 1. Calculate the column check code of the ECC code of the written data, and the column check code of the ECC code of the original data.

For example, it is assumed that the written data is 0xF0, the original data is 0x31, Data_ECC_CP indicates the column check code of the ECC code of the written data, and Buf_ECC_CP indicates the column check code of the ECC code of the original data, and the calculation of the above ECC code is performed. Method, you can calculate Data_ECC_CP={0,0,0,0,0}, Buf_ECC_CP={0,1,0,1,1,0}.

Step 2, according to the formula

A column check code of an ECC code of total data in a memory block to which the storage space to which the one-byte data is written is calculated.

among them,

A column check code indicating an ECC code of the modified total data, ECC_CP indicates a column check code of the ECC code of the total data before the modification, and ^ indicates a bitwise exclusive OR operation.

For example: suppose Data_ECC_CP={0,0,0,0,0}, Buf_ECC_CP={0,1,0,1,1,0},

then

Here is a brief introduction to the principle of the above formula: By analyzing the calculation method of the ECC code, the ECC_CP can be expressed in the form of ECC_CP=Byte0_ECC_CP^Byte1_ECC_CP^...^Byte255_ECC_CP, which is simplified to ECC_CP=X^Byte9_ECC_CP=X ^Buf_ECC_CP (assuming Byte 9 is the storage space to be written), and since ECC_CP^Buf_ECC_CP=X^Buf_ECC_CP^Buf_ECC_CP=X, then

visible,

The data of Byte9 before and after writing has been updated.

Secondly, the method for modifying the row check code of the ECC code may be: modifying the row check code of the ECC code of the total data in the storage block to which the storage space belongs according to the written data and the original data. The specific implementation process is:

Step 1. Determine the parity of the written data and the parity of the original data.

In the specific implementation process, the bit bits of the two data are XORed respectively. If the XOR result is 0, the parity is even. If the XOR result is 1, the parity is odd. For example: suppose the data written is 0xF0 (={1,1,1,1,0,0,0,0}), the original data is 0x31 (={0,0,1,1,0,0,0 , 1}), then for 0xF0, the XOR result is 1^1^1^1^0^0^0^0=0, its parity is even, for 0x31, the XOR result is 0^0^1^ 1^0^0^0^1=1, whose parity is odd.

In step 2, it is judged whether the parity of the written data is the same as the parity of the original data.

Step 3: If they are the same, the row check code of the ECC code of the total data is not modified; if different, the bit associated with the data in the byte to be written is determined in the row check code of the ECC code of the total data. Bit and reverse it.

On the one hand, if the parity of the written data is the same as the parity of the original data, by analyzing the calculation method of the ECC code, it can be known that the change of the written data and the original data does not affect the row check code of the ECC code, so It is not necessary to modify the row check code of the ECC code.

On the other hand, if the parity of the written data is not the same as the parity of the original data, the change of the written data and the original data may affect the row check code of the ECC code, so the line correction of the ECC code is to be modified. Code verification. In the specific implementation process, first determine the bit associated with the data in the written storage space in the row check code of the ECC code corresponding to the 256-byte data, wherein the association relationship can be obtained by referring to Table 2. For example, the bit associated with Byte0 in the row check code of the ECC code is RP0, RP2, RP4, RP6, RP8, RP10, RP12, RP14, and RP16; then the determined bit is inverted, where the inversion is Refers to changing data 0 to data 1, or changing data 1 to data 0. For example, suppose ECC_RP indicates the row check code of the ECC code before modification, ECC_RP={RP15, RP14, ..., RP3, RP2, RP1, RP0}={1,0,0,0,1,1,0,1,1,1,0,0,1,1,0,1},

Indicates the row check code of the modified ECC code. The associated bits are RP0, RP2, RP4, RP6, RP8, RP10, RP12, and RP14.

S104, according to the column check code of the ECC code of the total data in the storage block to which the storage space belongs after the one-byte data is written, and the ECC of the total data in the storage block to which the modified storage space belongs according to the modification The row check code of the code obtains the ECC code of the total data in the storage block to which the updated storage space belongs.

Specifically, since the ECC code includes the column check code and the row check code, the ECC code of the total data in the memory block after writing the data of one byte and the line of the ECC code of the total data in the modified memory block are The check code can obtain the ECC code of the total data in the storage block, and the ECC code is the updated ECC code.

It can be seen that, when the data is randomly written to the storage medium, the embodiment of the present invention only needs to modify the original ECC code, and the prior art must write 256 bytes of data to the storage block. The problem of calculating the ECC code is such that the random writeability of the storage medium is not destroyed. In addition, the embodiment of the present invention does not need to recalculate the ECC code, thereby improving the operation efficiency.

2 is a schematic flowchart of a method for verifying an ECC code according to an embodiment of the present invention. The method is applied to a data reading process of a storage medium, where the storage medium includes a plurality of storage blocks, and the storage block is A storage space composed of a plurality of bytes. Optionally, in the embodiment of the present invention, the storage block is set to a storage space of 256 bytes. The flow of the method for verifying the ECC code in this embodiment as shown in the figure may include:

S201. Perform parity check on the read data when the data is randomly read to the storage medium.

Specifically, the method flow for performing parity check on the read data is:

Step 1. Calculate the parity of the read data.

In the specific implementation process, the method of calculating the parity code is to obtain an XOR result for each bit bit of the read data. For example, if the read data is 0xF0 (={1,1,1,1,0,0,0,0}), the parity of 0xF0 is 1^1^1^1^0^0^ 0^0=0.

Step 2: Query the parity code of the storage space where the read data is stored when the data is written.

It should be noted that each byte in the storage medium, when writing data, calculates the parity code corresponding to the data in the storage space corresponding to the byte and saves it. In the specific implementation process, after reading the data, the parity code stored in the storage space where the read data is located is queried.

In step 3, it is judged whether the calculated parity code is the same as the saved parity code.

In step 4, if the result of the determination in step 3 is different, the parity error occurs; if the result of the determination in step 3 is the same, the parity is not in error.

On the one hand, if the calculated parity code is different from the saved parity code, it indicates that the stored data has a bit inversion, resulting in writing one data and reading another data. Step S202, to perform an ECC check.

On the other hand, if the calculated parity code is the same as the stored parity code, it means that the stored data does not have an abnormality, and the verification is completed, and it is not necessary to execute step S202. The ECC check needs to read 256 bytes of data in the entire storage block to calculate the ECC code, so that the check efficiency is low. Therefore, the embodiment of the present invention can avoid ECC check every time the data is read, and ensure that the ECC check is performed. The verification efficiency is improved while the verification efficiency is improved.

As an alternative implementation manner, the storage block is divided into a plurality of sub-memory blocks in advance, such as dividing a 256-byte storage block into eight 32-byte storage blocks, when randomly reading data from the storage medium, Perform parity check on the data in the sub-memory block where the read data is located. The parity check is similar to the above method. The only difference is that the parity code of the data in each byte is not saved, and 32 words are saved instead. The parity of the data in the section reduces the number of parity codes that need to be saved.

S202. Determine whether the parity is in error.

Specifically, if the parity is in error, step S203 is performed.

S203. Perform ECC code verification on the total data in the storage block to which the stored data of the read data belongs.

Specifically, the method for performing ECC code verification on the total data in the storage block to which the stored data belongs is:

Step 1: Calculate the ECC code of the total data in the storage block to which the storage space of the read data belongs.

In the specific implementation process, the storage block to which the stored data is located is determined, and then 256 bytes of data in the storage block are read, and the ECC code corresponding to the 256-byte data is calculated.

Step 2: Query the parity code saved when the storage block to which the read data belongs is written in the data.

It should be noted that each memory block in the storage medium calculates the parity code corresponding to the 256-byte data in the storage block when the data is written, and saves it. In the specific implementation process, after the data is read, the storage block to which the storage space in which the read data belongs is corresponding to the saved parity code.

Step 3, by comparing the calculated ECC code with the saved ECC code, determining a bit bit in the read data, and inverting the bit, wherein the inversion refers to changing the data 0 Is data 1, or data 1 is changed to data 0.

It should be understood that the ECC check can determine the bit of the error, and the specific determination method is not described herein. For example, if the read data is 0xF0 (={1,1,1,1,0,0,0,0}), it is determined that the bit inversion is Bit0, then the corrected data is 0xF1 (= {1,1,1,1,0,0,0,1}).

It can be seen that, when the data is randomly read from the storage medium, the embodiment of the present invention first performs parity check on the read data, and performs ECC check only when the parity error occurs, thereby avoiding reading data every time. ECC verification is required to ensure the verification reliability and improve the verification efficiency.

FIG. 3 is a schematic structural diagram of an apparatus for updating an ECC code according to an embodiment of the present invention. The apparatus of the present invention is for implementing the method of updating the ECC code provided in FIG. The device for updating the ECC code in the embodiment of the present invention may include at least an ECC code acquisition module 310, a data acquisition module 320, a row and column code modification module 330, and an ECC code update module 340, where:

The ECC code obtaining module 310 is configured to obtain an ECC code of total data in each of the storage blocks, where the ECC code includes a column check code and a row check code.

The data obtaining module 320 is configured to acquire, when the data is randomly written to the storage medium, the written data, the location of the storage space where the data is written, and the storage space is written in the Raw data stored before the data.

The location of the storage space is used to determine a storage block to which the storage space belongs. Optionally, the location of the storage space may be a storage address or a serial number of a byte. Take the serial number of the storage space as an example: assume that the data randomly written to the storage medium is 0xF0, and the byte corresponding to the storage space to which the data is written is Byte9, and the storage space is before the data is written. If the stored original data is 0x31, the obtained information is 0xF0, Byte9, and 0x31. Further, it is assumed that Byte0~Byte 255 belong to the first storage block, and since Byte9 falls in the interval of Byte 0 to Byte 255, it can be determined that the data is The written storage space belongs to the first storage block.

a lining code modification module 330, configured to pass a column check code of an ECC code of the written data, a column check code of an ECC code of the original data, and total data in a storage block to which the storage space belongs The column check code of the ECC code performs a logical operation to obtain a column check code of the ECC code of the total data in the memory block to which the storage space belongs after writing the one-byte data, and according to the written data and The original data modifies a row check code of an ECC code of total data in a storage block to which the storage space belongs. In a specific implementation, the row and column code modification module 330 may further include a column code calculation unit 331, a column code modification unit 332, a parity determination unit 333, and a row code modification unit 334 as shown in FIG. 4, wherein:

The column code calculation unit 331 is configured to calculate a column check code of the ECC code of the written data, and a column check code of the ECC code of the original data.

In the specific implementation process, the column check code of the ECC code of the written data is calculated, and the column check code of the ECC code of the original data; according to the formula

among them,

a column code modification unit 332 for

Calculating a column check code of an ECC code of total data in a storage block to which the storage space corresponding to the byte corresponding to the one byte of data is written, wherein

The parity determining unit 333 is configured to determine a parity of the written data and a parity of the original data.

In the specific implementation process, the bit bits of the two data are XORed respectively. If the XOR result is 0, the parity is even. If the XOR result is 1, the parity is odd.

The line code modification unit 334 is configured to determine whether the parity of the written data is the same as the parity of the original data, and if yes, the row check code of the ECC code of the total data is not modified; if not, Determining, in the row check code of the ECC code of the total data, a bit associated with the data in the written byte, and inverting it, wherein the inversion refers to the data 0 becomes data 1, or data 1 becomes data 0.

On the other hand, if the parity of the written data is not the same as the parity of the original data, the change of the written data and the original data may affect the row check code of the ECC code, so the line correction of the ECC code is to be modified. Code verification. In the specific implementation process, first determine the bit associated with the data in the written storage space in the row check code of the ECC code corresponding to the 256-byte data, wherein the association relationship can be obtained by referring to Table 2. For example, the bit associated with Byte0 in the row check code of the ECC code is RP0, RP2, RP4, RP6, RP8, RP10, RP12, RP14, and RP16; then the determined bit is inverted, where the inversion is Refers to changing data 0 to data 1, or changing data 1 to data 0.

The ECC code update module 340 is configured to: according to the column check code of the ECC code of the total data in the storage block to which the storage space belongs after the one-byte data is written, and the modified storage of the storage space The row check code of the ECC code of the total data in the block obtains the ECC code of the total data in the memory block to which the updated storage space belongs.

FIG. 5 is a schematic structural diagram of an apparatus for verifying an ECC code according to an embodiment of the present invention. The apparatus of the present invention is for implementing the method of verifying an ECC code provided in FIG. The device for verifying the ECC code in the embodiment of the present invention may include at least a parity check module 510, a check judging module 520, and an ECC code check module 530, where:

The parity check module 510 is configured to perform parity check on the read data when the data is randomly read to the storage medium. In a specific implementation, the parity check module 510 may further include a first calculating unit 511, a first query unit 512, and a code value determining unit 513 as shown in FIG. 6, wherein:

The first calculating unit 511 is configured to calculate a parity code of the read data.

In the specific implementation process, the method of calculating the parity code is to obtain an XOR result for each bit bit of the read data.

The first query unit 512 is configured to query a parity code saved when the read data is stored in the storage space.

It should be noted that each storage space in the storage medium calculates the parity code corresponding to the data in the storage space when the data is written, and saves it. In the specific implementation process, after the data is read, the storage space of the read data is queried corresponding to the saved parity code.

The code value determining unit 513 is configured to determine whether the calculated parity code is the same as the saved parity code; if not, the parity error occurs; if yes, the parity is not erroneous.

On the one hand, if the calculated parity code is different from the saved parity code, it indicates that the stored data has a bit inversion, resulting in writing one data, and reading another data, which is triggered at this time. The ECC code check module 530 performs ECC check.

On the other hand, if the calculated parity code is the same as the saved parity code, it means that the stored data does not have an abnormality, and the verification ends, and the ECC code verification module 530 does not need to be triggered. The ECC check needs to read 256 bytes of data in the entire storage block to calculate the ECC code, so that the check efficiency is low. Therefore, the embodiment of the present invention can avoid ECC check every time the data is read, and ensure that the ECC check is performed. The verification efficiency is improved while the verification efficiency is improved.

The verification determining module 520 is configured to determine whether the parity is in error.

The ECC code verification module 530 is configured to perform ECC code verification on the total data in the storage block to which the storage space of the read data belongs when the parity determination module determines that the parity error occurs. In a specific implementation, the ECC code verification module 530 may further include a second calculation unit 531, a second query unit 532, and a bit inversion unit 533 as shown in FIG. 7, wherein:

The second calculating unit 531 is configured to calculate an ECC code of the total data in the storage block to which the storage space of the read data belongs.

The second query unit 532 is configured to query a parity code that is saved when the storage block to which the read data belongs is written in the data.

a bit inverting unit 533, configured to determine, by comparing the calculated ECC code with the saved ECC code, a bit in which bit inversion occurs in the read data, and invert it, wherein The inversion refers to changing data 0 to data 1, or changing data 1 to data 0.

FIG. 8 is a schematic structural diagram of a storage device according to an embodiment of the present invention. As shown in FIG. 8, the storage device may include: at least one processor 801, such as a CPU, at least one communication bus 802, at least one storage medium 803, and a memory. 804, input and output interface 805. Among them, the communication bus 802 is used to implement connection communication between these components. The storage medium 803 includes a plurality of virtual storage blocks, and the storage block is a storage space of a plurality of bytes. The memory 804 can be a high speed RAM memory or can be Non-volatile memory, such as at least one disk storage. The input/output interface 805 is used to provide an interface for accessing an external device to read and write data. Optionally, the memory 804 may also be at least one storage device located away from the foregoing processor 801. A set of program codes is stored in the memory 804.

Wherein, in the data writing process of the storage medium 803, the processor 801 is configured to call the program code stored in the memory 804, and perform the following operations:

Optionally, the processor 801 passes the column check code of the ECC code of the written data, the column check code of the ECC code of the original data, and the ECC of the total data in the storage block to which the storage space belongs. The column check code of the code performs a logical operation, and the specific operation of the column check code of the ECC code of the total data in the storage block to which the storage space belongs after the data of one byte is written is:

Calculating a column check code of the ECC code of the written data, and a column check code of the ECC code of the original data;

According to the formula

Further, the specific operation of the processor 801 to modify the row check code of the ECC code of the total data in the storage block to which the storage space belongs according to the written data and the original data is:

Determining a parity of the written data and a parity of the original data;

Determining whether the parity of the written data is the same as the parity of the original data;

If yes, the row check code of the ECC code of the total data is not modified;

If not, determining a bit associated with the data in the written storage space in the row check code of the ECC code of the total data, and inverting, wherein the inversion is Refers to changing data 0 to data 1, or changing data 1 to data 0.

Optionally, the storage block is 256 bytes of storage space.

In addition, in the data reading process of the storage medium 803, the processor 801 is configured to call the program code stored in the memory 804, and performs the following operations:

Parsing the read data when the data is randomly read to the storage medium through the input/output interface;

Determining whether the parity is erroneous;

If yes, the ECC code is checked for the total data in the storage block to which the stored data of the read data belongs.

Optionally, the specific operation of the processor 801 for performing parity check on the data in the storage space where the read data is located is:

Calculating a parity code of the read data;

Querying a parity code stored in the storage space of the read data when the data is written;

Determining whether the calculated parity code is the same as the saved parity code;

If not, the parity error occurs; if so, the parity is not erroneous.

Optionally, the specific operation of the processor 801 to perform ECC code verification on the total data in the storage block to which the read data belongs is:

Calculating an ECC code of the total data in the storage block to which the storage space of the read data belongs;

Querying a parity code saved when the storage block to which the read data belongs is written in the data;

Determining a bit inversion in the read data by comparing the calculated ECC code with the saved ECC code, and inverting the bit, wherein the inversion means Data 0 becomes data 1, or data 1 becomes data 0.

Optionally, the storage block is 256 bytes of storage space.

One of ordinary skill in the art can understand that all or part of the process of implementing the foregoing embodiments can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium. When executed, the flow of an embodiment of the methods as described above may be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

The above is only the preferred embodiment of the present invention, and the scope of the present invention is not limited thereto, and thus equivalent changes made in the claims of the present invention are still within the scope of the present invention.

Claims

A method for updating an error detection and correction ECC code, the method being applied to a data writing process of a storage medium, the storage medium comprising a plurality of storage blocks, the method comprising:

Obtaining an ECC code of total data in each of the storage blocks, where the ECC code includes a column check code and a row check code;

When a byte of data is randomly written to the storage medium, the written data, the location of the storage space where the data is written, and the original data stored by the storage space before the data is written are acquired. Wherein the location of the storage space is used to determine a storage block to which the storage space belongs;

And performing, by using a column check code of the ECC code of the written data, a column check code of the ECC code of the original data, and a column check code of an ECC code of the total data in the storage block to which the storage space belongs. a logical operation of obtaining a column check code of an ECC code of total data in a storage block to which the storage space belongs after writing one-byte data, and modifying the data according to the written data and the original data a row check code of the ECC code of the total data in the storage block to which the storage space belongs;

a column check code of the ECC code of the total data in the memory block to which the storage space belongs after the one-byte data is written, and an ECC code of the total data in the memory block to which the modified storage space belongs The check code is obtained, and the ECC code of the total data in the storage block to which the updated storage space belongs is obtained.
The method according to claim 1, wherein said column check code of an ECC code of said written data, a column check code of an ECC code of said original data, and said storage space belong to The column check code of the ECC code of the total data in the storage block is logically operated, and the column check code of the ECC code of the total data in the storage block to which the storage space belongs after the one-byte data is written is obtained, including:

Calculating a column check code of the ECC code of the written data, and a column check code of the ECC code of the original data;

According to the formula
Calculating a column check code of an ECC code of total data in a storage block to which the storage space belongs after writing one-byte data, wherein
a column check code indicating an ECC code of the calculated total data, the ECC_CP indicating a column check code of an ECC code of the total data before calculation, and the Data_ECC_CP indicating an ECC code of the written data Column check code, the Buf_ECC_CP represents a column check code of the ECC code of the original data, and the ^ represents a bitwise exclusive OR operation.
The method according to claim 1, wherein said modifying a row check code of an ECC code of total data in a storage block to which said storage space belongs according to said written data and said original data, including :

Determining a parity of the written data and a parity of the original data;

Determining whether the parity of the written data is the same as the parity of the original data;

If yes, the row check code of the ECC code of the total data is not modified;

If not, determining a bit associated with the data in the written storage space in the row check code of the ECC code of the total data, and inverting, wherein the inversion is Refers to changing data 0 to data 1, or changing data 1 to data 0.
The method of any of claims 1-3, wherein the memory block is 256 bytes of storage space.
An apparatus for updating an error detection and correction ECC code, wherein the apparatus is applied to a data writing process of a storage medium, the storage medium comprising a plurality of storage blocks, the apparatus comprising:

An ECC code obtaining module, configured to acquire an ECC code of total data in each of the storage blocks, where the ECC code includes a column check code and a row check code;

a data acquisition module, configured to acquire, when the data is randomly written to the storage medium, the written data, the location of the storage space where the data is written, and the storage space in which the data is written Raw data stored previously, wherein the location of the storage space is used to determine a storage block to which the storage space belongs;

a row and column code modification module, configured to pass a column check code of an ECC code of the written data, a column check code of an ECC code of the original data, and an ECC of total data in a storage block to which the storage space belongs The column check code of the code performs a logical operation to obtain a column check code of the ECC code of the total data in the memory block to which the storage space belongs after writing the one-byte data, and according to the written data and the Determining the original data, modifying a row check code of the ECC code of the total data in the storage block to which the storage space belongs;

An ECC code update module, configured to store the one-byte data according to the storage space a row check code of the ECC code of the total data in the storage block of the genus and a row check code of the ECC code of the total data in the storage block to which the modified storage space belongs, and the updated storage of the storage space is obtained. The ECC code of the total data in the block.
The apparatus according to claim 5, wherein the row and column code modification module comprises:

a column code calculation unit, configured to calculate a column check code of the ECC code of the written data, and a column check code of the ECC code of the original data;

Column code modification unit for formulating
Calculating a column check code of an ECC code of total data in a storage block to which the storage space belongs after writing one-byte data, wherein
a column check code indicating an ECC code of the calculated total data, the ECC_CP indicating a column check code of an ECC code of the total data before calculation, and the Data_ECC_CP indicating an ECC code of the written data Column check code, the Buf_ECC_CP represents a column check code of the ECC code of the original c data, and the ^ represents a bitwise exclusive OR operation.
The apparatus according to claim 5, wherein the row and column code modification module comprises:

a parity determining unit, configured to determine a parity of the written data, and a parity of the original data;

a line code modification unit, configured to determine whether a parity of the written data is the same as a parity of the original data;

If yes, the row check code of the ECC code of the total data is not modified;

If not, determining a bit associated with the data in the written storage space in the row check code of the ECC code of the total data, and inverting, wherein the inversion is Refers to changing data 0 to data 1, or changing data 1 to data 0.
The apparatus according to any one of claims 5 to 7, wherein the memory block is a storage space of 256 bytes.