Western Digital Technologies, Inc.

Sandisk Technologies Inc.

Mahim Raj Gupta

Aaron Lee

Piyush Girish Sagdeo

Lior Avital

Ramkumar Subramanian

Rohan Dhekane

Alexander Bazarsky

Niles Yang

Xiying Costa

Bo Lei

Rohit Sehgal

Tomer Tzvi Eliash

Joanna Lai

Michael Micha Ionin

Mohsen Purahmad

Lola Grin

Itay Busnach

Leeladhar Agarwal

Lior Bublil

Sugandha Sharma

Daivik Bychapur Manjunatha

A block of memory in a data storage device can experience a read disturb based on how many times the block of memory was read, as well as on the program-erase count of the block. Non-data wordline maintenance triggered by program-erase count can address the read disturb in some situations but not in others, such as when a host is very read intensive and does not write to the memory that often. In one embodiment, maintenance on a non-data wordline in a block of memory is performed in response to the lifetime total amount of data read from the block being above a threshold. In another embodiment, a block is allocated for use based on the lifetime read count of the block and not only on the program-erase count of the block. Other embodiments are disclosed.

Data storage device and method for handling lifetime read disturb

Aspects of a storage device are provided for handling GBBs with defects between global interconnects and local word lines. The storage device includes a plurality of blocks each including a plurality of word lines, a plurality of interconnects within an interconnect set and between multiple ones of the blocks, volatile memory, and a controller. The controller determines a logical address pattern associated with the multiple ones of the blocks respectively including a program failure, determines whether the logical address pattern is associated with the interconnect set for the multiple ones of the blocks, determines a common word line associated with the program failures in the multiple ones of the blocks, un-marks these blocks as GBBs, and refrains from programming the common word line during respective program operations in the multiple ones of the blocks. Thus, blocks may be reclaimed, defective word line(s) isolated, and likelihood of read-only modes reduced.

System recovery during CGI-WL defect

Technology is disclosed herein for checking data integrity in a non-volatile storage system. The storage system may operate in a first mode in which a data integrity check is performed in closed blocks until more than an allowed number of word lines fail the data integrity check. After a closed block has more than the allowed number of the word lines fail the data integrity check, then the storage system may operate in a second mode in which a data integrity check is performed in open blocks. The allowed number of word lines may be equal to the number of word lines that can be recovered by XOR data in the event data is uncorrectable by an ECC engine. The data integrity check of a target word line in an open block may be performed after programming a word line adjacent to the target word line in the open block.

Data integrity check in non-volatile storage

A data storage device includes a memory device and a controller coupled to the memory device. The controller is configured to determine that a power loss event has occurred, determine that one or more blocks are in an erased state, examine a block of the one or more blocks to determine whether the block is a SLC erased block or a TLC erased block, and place the block in a SLC pre-erase heap if the block is the SLC erased block or in a TLC pre-erase heap if the block is the TLC erased block. The controller is further configured to determine a first bit count of page0 for a SLC voltage for the block, determine a second bit count of page1 for a TLC voltage for the block, and classify the block as either a SLC erased block or a TLC erased block.

Block erase type detection using bit count check

A data storage device includes a memory device having a plurality of blocks and a controller coupled to the memory device. The controller is configured to determine that an uncorrectable error correction code (UECC) failure has occurred to a block of the plurality of blocks, enable a UECC anti-strike mechanism, and erase the block. The UECC anti-strike mechanism comprises converting a read failure associated with the block to an erase failure. The controller is further configured to retire the block upon determining that the erase is unsuccessful.

UECC failure handling method

Identifying a faulty memory die in a non-volatile memory storage system. Various methods include: commencing a programming operation of a multi-state block of a subject die, the programming operation including populating a transfer data latch with a first set of data and transferring the data to a first data latch, populating the transfer data latch with a second set of data and transferring the data to second data latch, arranging the first and second data sets in a suitable format of the multi-state block, and writing the data sets to the multi-state block; prior to populating the transfer data latch with the second data set, performing a program suspend and read operation thereby populating the transfer data latch with read data; and comparing the read data to the data contained in the first data latch and, if the comparison results in a match, identifying the subject die as faulty.

System countermeasure for read operation during TLC program suspend causing ADL data reset with XDL data

Detecting a faulty memory block. Various methods include: performing a read operation on a memory block of the memory array, the read operation generates a failed bit count; determining the failed bit count in above a value associated with an overall failed bit count; determining the failed bit count is above a threshold value; in response, performing a confirmation process on the memory block, the confirmation process defining a number of consecutive erase cycles and a level of an erase cycle, the confirmation process results in erase pass or erase fail; and marking the memory block for garbage collection in response to determining the confirmation process results in erase fail. Methods additionally include setting the level of the erase cycle by modifying at least one selected form the group comprising: an erase voltage parameter; an erase verify parameter; and a number of bits ignored during the erase cycle.

Detecting a faulty memory block

A non-volatile storage system comprises memory cells arranged in groups of memory cells that include programmable select gates and one or more control circuits in communication with the memory cells. The one or more control circuits configured to identify a select gate that needs to be programmed and program the select gate identified to be programmed if a temperature at the non-volatile memory cells is greater than a minimum temperature and defer programming of the select gate identified to be programmed until the temperature at the non-volatile memory cells is greater than the minimum temperature. In some embodiments, the one or more control circuits are configured to perform dummy memory operations on the plurality of non-volatile memory cells to raise the temperature of the non-volatile memory cells in response to determining that the temperature at the non-volatile memory cells is not high enough.

Growing community of inventors

Milpitas, CA, United States of America

Mahim Raj Gupta