System protection method for abnormal bit flipping of Nand Flash

Abstract

The invention discloses a system protection method for abnormal bit flipping of a Nand Flash, which can solve the problem that when bit flipping occurs in a blank area of a physical block of the Nand Flash in a severe environment, a UBI file system detects the bit flipping but cannot be repaired by ECC (Error Correction Code), so that the system enters an exception handling process and is forcibly changed into read-only protection. After the UBI file system detects that bit inversion occurs in the blank area of the physical block of the Nand Flash, 0 * ff is rewritten in the blank area so as to repair the blank area, data in a data area is carried into a standby block, an original block is erased, whether the data is erased completely or not is checked, and an additional read-write test is carried out. And marking the block as a bad block or putting the block into a standby block list according to whether the block can be erased or not and a read-write test result. And meanwhile, the mapping between the logic block and the original physical block subjected to bit flipping is canceled, and the logic block is remapped to the standby block which obtains the migrated data, so that the continuity of the storage space of the file system is realized, and the starting and normal operation of the Nand Flash-based system are ensured.

Description

Technical Field

[0001] The invention relates to the technical field of embedded devices, and specifically designs a system protection method for Nand Flash abnormal bit flipping. Background Art

[0002] Nand Flash is a commonly used storage medium for embedded devices. It generally has the problem of bit flipping. Nand Flash itself provides ECC check at the driver layer to correct a certain degree of bit flipping. When writing data to a Nand Flash page, an original ECC checksum will be generated at the same time. When reading the data of this page later, this ECC data will be used to check and repair the data errors. Therefore, since the blank page has no data written, ECC data will not be stored. However, under some extreme conditions, these blank pages of Nand Flash data will also have bit flipping. When reading data from Flash, the UBI file system based on Nand Flash will detect the valid data area and the data blank area at the same time. When the data blank area is detected to have bit flipping, there is no corresponding ECC code to correct the data, so the UBI file system will enter the exception handling process and force the system to switch to read-only mode for protection. At this time, the normal write operation to Nand Flash will fail, resulting in system abnormality. Therefore, how to effectively solve the various impacts that Nand Flash bit flipping may cause to the system has also become a major issue in the application of Nand Flash in embedded systems. Summary of the invention

[0003] The present invention provides a system protection method for Nand Flash abnormal bit flipping, which can effectively solve the problems in the background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] A system protection method for Nand Flash abnormal bit flipping includes the following steps:

[0006] S100, after the UBI file system detects that a bit flip occurs in a data blank area of ​​a physical block of the Nand Flash, the number of bit flips is counted and all blank areas of the physical block where the bit flip occurs are set to 0xff, thereby correcting the bit flip;

[0007] S200, moving the data of the physical block (including the valid data area and the corrected data blank area) to a spare block;

[0008] S300, erase the original block, check whether it is completely erased and perform additional read / write tests. If it has been erased and the read / write tests are normal, the physical block is put into the spare block list for subsequent use. If it is detected that it cannot be completely erased or the bit flip problem still occurs during the read / write test, the block is marked as a bad block at the MTD layer and stored in the bad block table to avoid next use;

[0009] S400, cancel the mapping between the logic block and the original bit-flipped physical block, enable the spare block for moving data, and remap the original corresponding logic block to the spare block, thereby achieving the continuity of the file system storage space.

[0010] It can be seen from the above technical solution that the system protection method for Nand Flash abnormal bit flipping of the present invention has the following features:

[0011] Beneficial effects:

[0012] When a bit flip occurs in a blank area of ​​a physical block of Nand Flash in a harsh environment, the problem block can be effectively processed by actively correcting the bit flip and moving data, thereby avoiding the failure of the file system entering the exception handling program and the system being forced to change to read-only mode due to the lack of ECC data for repair. The present invention effectively protects the system based on Nand Flash storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0014] Figure 1 is a flow chart of the present invention;

[0015] Figure 2 It is a schematic diagram of the method of the present invention; DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] like Figure 1 As shown, the system protection method for Nand Flash abnormal bit flipping described in this embodiment includes the following steps:

[0018] S100, after the UBI file system detects that a bit flip occurs in a data blank area of ​​a physical block of the Nand Flash, the number of bit flips is counted (because the flips in the blank area are all 1 to 0) and the blank area of ​​the physical block where the bit flip occurs is rewritten with 0xff, thereby correcting the bit flip;

[0019] S200, moving the data of the physical block (including the valid data area and the corrected data blank area) to a spare block;

[0020] S300, erase the original block, check whether it is completely erased and perform additional read / write tests. If it has been erased and the read / write tests are normal, the physical block is put into the spare block list for subsequent use. If it is detected that it cannot be completely erased or the bit flip problem still occurs during the read / write test, the block is marked as a bad block at the MTD layer and stored in the bad block table to avoid next use;

[0021] S400, cancel the mapping between the logic block and the original bit-flipped physical block, enable the spare block for moving data, and remap the original corresponding logic block to the spare block, thereby achieving the continuity of the file system storage space.

[0022] Figure 2 It is the principle diagram of the method of the present invention, and the following is combined with Figure 2 The operating principle of the embodiment of the present invention is described in detail:

[0023] When the UBI file system detects that a bit flip occurs in a data blank area of ​​a physical block when reading Nand Flash data, the present invention does not enter an exception processing flow even if there is no ECC data for correction. Instead, the present invention utilizes the characteristic that the data in the blank area is all 0xff, that is, the bit flips are all flipped from 1 to 0, and can rewrite 0xff in the non-0xff area for correction.

[0024] Since it is uncertain whether the flip is accidental or permanent and irreparable, that is, it is uncertain whether the block is usable after correction, it is necessary to move the repaired block data to the spare block completely to avoid data loss, and then perform an erase test on the original block. It should also be noted that if the original block has an irreparable bit flip, the operation of rewriting 0xff to the original block may fail, but the original block data that can still be repaired will be stored in the spare block, because the file system will still perform the same detection and repair operation when remapping and reading the spare block data later, and the flipped bit data will be repaired in the spare block.

[0025] Re-erase the original flipped block and perform additional read and write tests to evaluate the availability of this block. If it can be completely erased and the test data is read and written normally, it is considered that an accidental bit flip occurred, and the block is marked as a spare block and put into the spare block list, waiting to be reused when the UBI file system performs load balancing or bad block processing. If it is detected that the data cannot be erased or the read and write data verification is abnormal, it is considered that this block has a permanent and unrepairable bit flip problem, and it is necessary to mark this block as a bad block at the MTD layer, store it in the bad block list and no longer use it.

[0026] Finally, the logical mapping between the UBI file system logical block and the original flipped block is canceled, and the spare block for the moved data is enabled, and the original corresponding logical block is remapped to this spare block. In this way, the original data is preserved and the flipped block is replaced, ensuring the continuity of the file system storage space.

[0027] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A system protection method for Nand Flash abnormal bit flipping, characterized by: The following steps are involved: S100, after the UBI file system detects that a bit flip occurs in a data blank area of ​​a physical block of the Nand Flash, the number of bit flips is counted and the blank area of ​​the physical block where the bit flip occurs is rewritten with 0xff to correct the bit flip; S200, moving the data of the physical block (including the valid data area and the corrected data blank area) to a spare block; S300, erase the original block, check whether it is completely erased and perform additional read / write tests. If it has been erased and the read / write tests are normal, the physical block is put into the spare block list for subsequent use. If it is detected that it cannot be completely erased or the bit flip problem still occurs during the read / write test, the block is marked as a bad block at the MTD layer and stored in the bad block table to avoid next use; S400, cancel the mapping between the logic block and the original bit-flipped physical block, enable the spare block for moving data, and remap the original corresponding logic block to the spare block, thereby achieving the continuity of the file system storage space.

Images