JP2010092574A - Flash file system error correction function - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that in a flash file system, an error correction means is indispensable when data is read from and written to a flash memory, and high computational complexity is required when an error correction function is used. <P>SOLUTION: Only error detection is performed before error correction. Appropriate error correction is performed only when there is an error. The number of error detections and the number of error corrections of the error correction means are varied, depending on the type of a program or data on the flash memory. As a result, the processing time of error correction is reduced. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an error correction function of a flash file system, which relates to performing error correction after error detection, or changing the ability of the error correction function depending on the type of program on a flash memory.

  Recently, the capacity of external storage devices has increased exponentially and the transfer time has also increased. Therefore, we have taken measures by increasing the transfer speed. However, when the transfer speed is increased, data transfer errors increase, and an error correction system has become indispensable. Against this background, many types of error correction systems have been proposed, how many bits of error can be detected and how many bits of error can be corrected, and how many error correction bits can be added for that purpose. There are appropriate error correction algorithms depending on the number of error detections and the number of error corrections.

  In flash file systems using NAND flash memory, error correction has become indispensable as memory capacity increases and read / write time decreases. In recent years, NAND flash memory has increased in NAND flash memory with multiple values, the data error rate has increased, and error correction means has become an even more important factor. On the other hand, the error correction means uses Hamming error correction with a Hamming distance of 3, but BCH error correction has become necessary as the number of error detection and error correction increases. BCH error correction creates an error correction bit using information bits and a generator polynomial, adds it to the information bit, writes it to the flash memory, and when reading from the flash memory, calculates the syndrome after reading it to determine whether there is an error If there is an error, the error was corrected again by calculation or a lookup table.

  Specifically, as shown in FIG. 4, I (x): information bit is divided by G (x): generator polynomial to obtain h (x): quotient and R (x): remainder. R (x): the remainder is an error correction bit, I (x): data after error correction with I (x) + R (x) after the information bit. After writing to and reading from the flash memory in this form, one syndrome of α * 4 + α + 1 = 0 is substituted into x of I (x) + R (x) to obtain a syndrome of S (α). S (α): If the syndrome is all zero, there is no error. However, if S (α) is not 0, an error has occurred. Therefore, the error is corrected using an error correction calculation or a lookup table.

Although there is no literature regarding the error correction function of the flash file system, the following is related to the error correction of the file system.
JP 2000-011689 A JP-A-11-272568 JP-A-11-143787 JP-A-11-232778

  As described above, the error correction process consumes a great amount of time such as calculation by a generator polynomial, and the process of correcting an error from the syndrome also consumes a lot of time. That is, the error correction work for each reading and writing is a heavy process for the flash file system. That is, increasing the number of error corrections and error detections increases the processing time.

  Therefore, when calculating a predetermined number of error correction bits capable of error detection and error correction using a generator polynomial, adding them to information bits and writing them to the flash memory, first, only error detection is performed with the maximum number of error detections, Based on the number of detected errors, correct the error again, or change the error detection number and error correction number of the error correction means that reads and writes the program and data on the flash memory according to the type of program and data on the flash memory. Or, depending on the storage location of the program or data on the flash memory, change the error detection number and error correction number of the error correction means for reading and writing the program or data on the flash memory, or change the program or data on the flash memory to Depending on the added flag, the program or data on the flash memory It changed the error count and the error correction number of the error correction means to read and write.

  In general, it takes a lot of processing time for error correction means.By shortening the time of error correction processing, such as detecting whether there is an error first or reducing the error detection function according to the importance of the program, The system has become very efficient.

  Taking an example of error correction detection means having an error correction function with a minimum distance of 6 or more, 5-bit error detection can be performed when error correction processing is not performed. When 1-bit error correction is performed from the syndrome, 4-bit error detection can be performed. When 2-bit error correction is performed from two syndromes, 3-bit error detection can be performed. From this, first, 5-bit error detection is performed, and after determining the number of errors, appropriate error correction is performed. As a result, the maximum number of error detection functions can be used to avoid error detection, and then a fixed number of error detections can be obtained and appropriate error correction can be performed.

  In FIG. 1, 1 is a flash memory space. In the boot code storage area 2, for example, an error correction function capable of performing error 3 bit correction and 4 bit detection is added and stored. In the OS storage area 3, for example, an error correction function capable of performing error 2-bit correction and 3-bit detection is added and stored. The user data storage area 4 stores, for example, an error correction function capable of error 1it correction and 2-bit detection.

  By changing the capability of the error correction function depending on the type of program, the error correction function with the highest error correction capability is attached to the boot code, and then the error correction function with the highest error correction capability is attached to the OS. Finally, an error correction function with the lowest error correction capability was attached to user data. In this way, a method having a high error correction function is added to those that should not have an error even if processing time is required, and a method in which processing time is given priority to those that do not.

  In FIG. 2, 1 is a flash memory space. The boot code of No. 2 is stored in the first block of the flash memory with an error correction function that can detect, for example, 3-bit correction and 4-bit detection. The first block of the flash memory can be stored continuously because there is no congenital defective block. For example, the OS of 3 is provided with an error correction function capable of performing error 2-bit correction and 3-bit detection, and stored in a predetermined block. For example, the user data 4 is stored with an error correction function capable of error 1it correction and 2-bit detection.

  In this way, the type of program is stored in a predetermined block, and the error correction function with the highest error correction capability is changed for each block. Next, the error correction function with the highest error correction capability is assigned to the block in which the OS is stored, and finally the error correction function with the lowest error correction capability is assigned to the block in which the user data is stored. It was. By doing this, even if processing time is required for those that should not have errors, those with high error correction function are added and assigned to specific blocks, and for those that do not, processing time has priority. Assigned to a specific block to do. Therefore, the type of program can be specified by the type of block, and an error correction function having free error correction capability can be added.

  In FIG. 3, 1 is a flash memory space. The boot code of No. 2 is provided with an error correction function that can detect, for example, error 3 bit correction and 4 bit detection, and a flag 1 that identifies the boot code is attached and stored in the flash memory. For example, an error correction function capable of performing error 2 bit correction and 3 bit detection is added to the OS 3, and a flag 2 that identifies the OS is added and stored in the flash memory. For example, an error correction function capable of error 1it correction and 2-bit detection is added to the user data No. 4 and a flag 3 that identifies user data is added to the flash memory.

  In this way, the type of program can be recognized with a predetermined flag, and the error correction function with the highest error correction capability is attached to the boot code by changing the error correction function for each program type. It is stored in the flash memory together with the flag 1, and then the error correction function having the highest error correction capability is attached to the OS, and stored in the flash memory together with the flag 2, and finally the error correction function having the lowest error correction capability is stored in the user data. And stored in the flash memory together with the flag 3. In this way, even if processing time is required, an error correction having a high error correction function is added to an object that should not have an error, and its presence is recognized by a flag. In addition, the type of program can be specified by attaching a specific flag to a method that gives priority to the processing time, and an error correction function having a free error correction capability can be added.

  In general, it takes a lot of processing time for error correction means.By shortening the time of error correction processing, such as detecting whether there is an error first or reducing the error detection function according to the importance of the program, The industrial availability of the flash file system has become very high.

It is a block diagram of arrangement | positioning of a boot code, OS, and user data on flash memory space. It is a block diagram of arrangement | positioning of a boot code, OS, and user data on flash memory space. It is a block diagram of arrangement | positioning of a boot code, OS, and user data on flash memory space. It is a figure about the method of making the error correction bit of an error correction function, and the error detection and correction method after reading.

Explanation of symbols

1 Flash memory space 2 Boot code storage area 3 OS storage area 4 User data storage area 5 Boot code 6 OS
7 User data

Claims (4)

  When a predetermined number of error correction bits capable of error detection and error correction are calculated using a generator polynomial, added to the information bits, written to the flash memory, and read out, the error detection is first performed with the maximum number of error detections detected. A flash file system that performs error correction again based on the number of errors.   In a flash file system with NAND flash memory, the error detection number and error correction number of the error correction means for reading and writing the program and data on the flash memory are changed depending on the type of program and data on the flash memory. And a flash file system.   In a flash file system with NAND flash memory, the error detection number and error correction number of error correction means for reading and writing the program and data on the flash memory can be changed depending on the storage location of the program and data on the flash memory. A featured flash file system.   In a flash file system with NAND flash memory, the error detection number and error correction number of the error correction means for reading and writing the program and data on the flash memory can be changed according to the flag added to the program and data on the flash memory. Flash file system featuring.

Images