Method and device for generating trial and error table of nonvolatile memory

Abstract

The invention discloses a method and a device for generating a trial and error table of a nonvolatile memory. The method comprises the steps of obtaining an environment variable causing a read data error of the nonvolatile memory; determining a quantitative description value of the environment variable, wherein the quantitative description value is used for describing the environment variable through accurate data; reading the data of the nonvolatile memory under the quantitative description values of different environment variables respectively; and generating a trial and error table of the nonvolatile memory according to the data read from the nonvolatile memory. According to the method and the device, the technical problems of relatively low error correction efficiency and relatively few use scenes covered by a fixed trial and error table provided by a manufacturer due to the fact that a fixed trial and error table provided by a nand flash manufacturer is used for correcting read error data when UECC (Unified Error Correction Code) occurs during reading of the Nand flash are solved.

Description

Technical field[0001]The present application relates to the field of storage devices, and in particular to a method and device for generating a trial and error table of a non-volatile memory.Background technique[0002]When UECC (Uncorrectable Error Checking and Correcting) appears when reading Nand flash, the first step is to use the retry table to correct, and to correct the read error data by trying different retry table values.[0003]The existing retry method is to query the fixed retry table provided by the nand flash manufacturer, and use the entries provided by the retry table to try to read the nand flash in turn. If the data is corrected, stop the table lookup; if the data cannot be corrected, the retry table will be checked until the entire table is searched.[0004]The existing retry methods mainly have the following disadvantages:[0005]1) Because you need to query the retry table one by one, when the correction is not returned, the next entry in the table will be tried. This ...

AI Technical Summary

Problems solved by technology

[0005] 1) Because the retry table needs to be queried one by one, when the correction fails, the next entry in the table will be tried, which will inevitably increase the number of trial and error and increase the delay in processing write commands, so that the solid state disk (Solid State Disk, SSD) processes data would become insanely slow, giving customers a poor user experience

[0006] 2) The retry table provided by the nand flash manufacturer is only the retry value under some specific environment variables, and the usage scenarios of SSD are ever-changing. Many scenarios cannot be covered by the retry table provided by the manufacturer, so the data may be corrected after all the table items are tried. If you don't come back, you will waste a lot of time processing orders

[0007] For the above problems, no effective solution has been proposed

Images