Read retry method for solid state storage device
A read retry method for a solid state storage device is provided. The solid state storage device is in communication with a host. The solid state storage device includes a non-volatile memory. The read retry method includes the following steps. Firstly, the solid state storage device judges whether a specified read block of the non-volatile memory is in a specified failure mode. If the specified read block of the non-volatile memory is in the specified failure mode, a failure mode read retry process corresponding to the specified failure mode is performed. If an accurate read data is acquired in the failure mode read retry process, the accurate read data is transmitted to the host. If the accurate read data is not acquired in the failure mode read retry process, a read fail message is sent to the host.
1. A method for recovering a solid state storage device from a memory accessing error, wherein the solid state storage device comprises a memory and a control circuit, and the control circuit comprises a failure mode judging circuit, the method comprising steps of:
judging whether the memory is in one of a first failure mode and a second failure mode by the failure mode judging circuit after the memory accessing error occurs on a specified read block, wherein the first failure mode associates with a first error handling operating sequence, the second failure mode associates with a second error handling operating sequence, and the first error handling operating sequence is different from the second error handling operating sequence; and
performing the first error handling operating sequence by the control circuit in response to the failure mode judging circuit determining the memory is in the first failure mode;
wherein the first failure mode is a write/read temperature difference failure mode and the first error handling operating sequence comprises steps of:
judging an erase count of the specified read block;
if the erase count is larger than a threshold value, performing a read level calibration process and then, if the read level calibration process is unsuccessful, performing a soft decoding process; and
if the erase count is not larger than the threshold value, performing a read level shift decoding process, and then, if the read level shift decoding process is unsuccessful, performing the read level calibration process, and then, if the read level shift decoding process and the read level calibration process are both unsuccessful, performing the soft decoding process.
2. The method as claimed in claim 1 , wherein the second failure mode is a data retention failure mode and the second error handling operating sequence includes another read level shift decoding process, another read level calibration process, and another soft decoding process.
3. The method as claimed in claim 1 , wherein the second failure mode is a read disturbing failure mode and the second error handling operating sequence includes another read level shift decoding process, another read level calibration process, a RAID recovery process, and another soft decoding process.
4. The method as claimed in claim 1 , further comprising:
performing a default error handling operating sequence by the control circuit in response to the failure mode judging circuit determining the memory is not in the first failure mode or in the second failure mode;
wherein the first error handling operating sequence associates with a first subset of a plurality of predefined sets of read retry voltages, and the default error handling operating sequence associates with the plurality of predefined sets of read retry voltages.
5. The method as claimed in claim 4 , wherein the default error handling operating sequence comprises another read level shift decoding process, another read level calibration process, a RAID recovery process, and another soft decoding process.