Bypass buffer for wear codewords

By designing a bypass buffer in the memory device to identify and process wear codewords, the problem of insufficient correction capability of ECC in the face of a large number of fixed faults is solved, and the reliability and life of the data storage device are improved.

CN119025017APending Publication Date: 2024-11-26SANDISK TECHNOLOGIES LLC
0 Cites 0 Cited by

Patent Information

Application Number
CN202410010475.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-07-26
Filing Date
2024-01-04
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In the face of a large number of fixed faults, it is difficult to effectively correct soft bit errors in memory devices, resulting in a decrease in the reliability and life of the data storage device.

Method used

A bypass buffer is designed to monitor the number of codeword wear bits in the memory device, store the relevant data and addresses in the bypass buffer when the threshold exceeds the threshold, and check the bypass buffer when the command is received, to Decide whether to perform a read or write operation to the codeword or address.

Benefits of technology

Improves the reliability of aging data storage devices, extends device life, and relaxes ECC calibration capability requirements, allowing resources to be used for other operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119025017A_ABST
    Figure CN119025017A_ABST
Patent Text Reader

Abstract

The present disclosure provides a bypass buffer that stores codewords that have been identified as worn codewords due to fixed-type fail bit errors and / or other persistent failures. A controller of a data storage device monitors a number of wear bits of a codeword stored by a memory device of the data storage device. If the number of worn bits of the codeword exceeds a threshold number of worn bits, data associated with the codeword is corrected and stored in the bypass buffer. A memory address associated with the codeword is also stored in the bypass buffer and associated with the corrected codeword.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Application No. 63 / 502,572, filed on May 16, 2023, entitled “BYPASSBUFFER FOR WORN OUT CODEWORDS,” the entire disclosure of which is hereby incorporated by reference into this document. Background Art

[0003] A stuck-at-failure is a type of persistent failure in memory devices including magnetic random access memory (MRAM) devices and phase change memory (PCM) devices. In general, a stuck-at-failure is a type of bit error that occurs when a specific memory cell of a memory device is fixed at a specific logic level (e.g., logic "1" or logic "0") regardless of an input signal and / or output signal associated with the specific memory cell.

[0004] As memory devices age, fixed-type faults may become one of the main types of failures in memory devices. Existing solutions include using strong error correction codes (ECC) to correct fixed-type failure bits of specific codewords. However, ECC can usually only correct a certain number of bits within a codeword. In addition, the bit correction capability of ECC is for all faults associated with the codeword (e.g., soft bit errors, such as retention errors and / or read interference errors), not just fixed-type faults. Therefore, when more fixed-type faults occur, the ability of ECC to correct other types of errors is significantly reduced.

[0005] Therefore, it would be beneficial to identify and bypass codewords stored in memory devices that have a large number of stuck-at failure bits. Summary of the invention

[0006] The present application describes a bypass buffer that stores codewords that have been identified as worn (or have a large number of worn bits) due to fixed-type fault bit errors and / or other persistent faults. As will be explained in more detail herein, a controller associated with a memory device can monitor the number of worn bits (e.g., bits of a codeword that are fixed at a specific logic level) of a codeword stored by the memory device. If the number of worn bits of a codeword exceeds a threshold number of worn bits, the data associated with the codeword is corrected and stored in a bypass buffer. A memory address associated with the codeword is also stored in a bypass buffer.

[0007] When a command (e.g., a read command or a write command) is received, the bypass buffer is checked to determine whether the memory address associated with the command is included as an entry in the bypass buffer. If the memory address associated with the command is included as an entry in the bypass buffer, the command is executed on the codeword / entry in the bypass buffer. However, if the memory address associated with the command is not found as an entry in the bypass buffer, the command can be executed on the data associated with the memory address of the memory device.

[0008] Therefore, the present application describes a method for identifying worn codewords. In one example, the method includes reading data associated with a codeword located at a memory address of a memory device. The data associated with the codeword is stored in a temporary storage location. Test data is written to the memory address associated with the codeword. In one example, the test data includes a plurality of bits, each of which has a first state. The test data is then read from the memory address associated with the codeword. The number of bits in the test data having a second state different from the first state is determined. Based at least in part on determining that the number of bits in the second state is above a threshold, the data associated with the codeword is stored as an entry in a bypass buffer. In addition, the entry in the bypass buffer is associated with the memory address of the memory device.

[0009] The present application also describes a data storage device including a controller and a memory. The memory is communicatively coupled to the controller and stores instructions to be executed by the controller. When the controller executes the instructions, the controller performs various operations. In one example, the operation includes reading data associated with a codeword at a memory address of a memory device located in the data storage device, and storing the data associated with the codeword in a temporary storage location. Test data is written to the memory address of the memory device to determine the number of wear bits associated with the codeword. In one example, the test data includes a plurality of bits, each of which is in the same state. The test data may then be read from the memory address. The number of bits with different states in the test data is determined. Based at least in part on determining that the number of bits in different states is above a threshold, the data associated with the codeword is stored as an entry in a bypass buffer, and the entry in the bypass buffer is associated with the memory address of the memory device.

[0010] A data storage device is also described. The data storage device includes: a device for reading data associated with a codeword at a memory address of a memory device of the data storage device; and a device for storing the data associated with the codeword in a temporary storage location. The data storage device may also include a device for writing test data to the memory address. The data storage device may also include a device for determining the number of wear bits associated with the codeword. In one example, the number of wear bits is based at least in part on the number of bits having a first state associated with the test data and the number of bits having a second state different from the first state. The data storage device also includes a device for storing the data associated with the codeword as an entry in a bypass buffer. In one example, when it is determined that the number of bits in the second state is above a threshold, the data associated with the codeword is stored as an entry in the bypass buffer. The data storage device also includes a device for associating the entry in the bypass buffer with the memory address of the memory device.

[0011] This Summary is provided to introduce some concepts in a simplified form that are further described in the Detailed Description below. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Non-limiting and non-exhaustive examples are described with reference to the following figures.

[0013] Figure 1 is a block diagram of a system including a host device and a data storage device according to one example.

[0014] Figure 2 A bypass buffer for storing worn codewords is shown according to one example.

[0015] Figure 3 A method for determining whether a codeword should be added to a bypass buffer according to one example is shown.

[0016] Figure 4 A method for performing a read operation on a bypass buffer according to one example is shown.

[0017] Figure 5 Another method for performing a read operation on a bypass buffer according to an example is shown.

[0018] Figure 6 A method for performing a write operation on a bypass buffer according to one example is shown.

[0019] Figure 7Another method for performing a write operation on a bypass buffer according to an example is shown.

[0020] Figure 8 A bypass buffer with a confidence indicator according to one example is shown.

[0021] Fig. 9 A method for determining whether a codeword should be added to a bypass buffer according to one example is shown.

[0022] Fig.10 is a perspective view of a memory device including a three-dimensional (3D) stacked nonvolatile memory according to one example.

[0023] Fig.11 is a block diagram of a storage device according to an example. DETAILED DESCRIPTION

[0024] In the following detailed description, reference is made to the accompanying drawings which form a part of the detailed description, and in which specific embodiments or examples are shown by way of illustration. These aspects may be combined, other aspects may be utilized, and structural changes may be made without departing from the present disclosure. Therefore, the following detailed description should not be taken in a limiting sense, and the scope of the present disclosure is defined by the appended claims and their equivalents.

[0025] As previously mentioned, a stuck-at fault is a type of persistent fault in many types of memory devices, including MRAM devices and PCM devices. In general, a stuck-at fault is a type of bit error that occurs when a specific cell (or cells) of a memory device is stuck at a logic "1" or logic "0" regardless of the signal associated with the specific memory cell.

[0026] When data storage devices age, fixed-type faults may become one of the main fault types. In order to solve the problem of fixed-type faults, strong error correction codes (ECC) are used. However, ECC can usually only correct a certain number of bits within a specific codeword. For example, a specific ECC can correct up to eleven bits of a specific codeword. However, the corrected eleven bits may include fixed-type fault errors and soft bit errors. Therefore, if a specific codeword has six fixed-type fault errors, ECC can only correct five soft errors. When more fixed-type faults occur, the ability to correct other types of errors is significantly reduced.

[0027] To address the above problems, the present application describes a bypass buffer that stores codewords that have been identified as worn due to fixed-type fault bit errors and / or other persistent faults. As will be explained in more detail herein, a controller of a data storage device monitors the number of worn bits (e.g., bits of a codeword that are fixed at a specific logic level) of codewords stored by a memory device of the data storage device. If the number of worn bits of a codeword exceeds a threshold number of worn bits, the data associated with the codeword is corrected and stored in a bypass buffer. A memory address associated with the codeword is also stored in a bypass buffer and associated with the codeword.

[0028] When a command (e.g., a read command or a write command) is received, the bypass buffer is checked to determine whether the memory address associated with the command is included as an entry in the bypass buffer. If the address associated with the command is included as an entry in the bypass buffer, the command is executed on the codeword / entry in the bypass buffer. However, if the address associated with the command is not found as an entry in the bypass buffer, the command may be executed on the data associated with the memory address of the memory device.

[0029] Thus, many technical advantages can be achieved, including but not limited to improving the reliability of aging data storage devices by helping to ensure that ECC capability is preserved against soft errors, extending the life of data storage devices, and relaxing ECC capability requirements of aging data storage devices so that resources can be utilized by other operations / tasks.

[0030] About Figures 1 to 11 These benefits and other examples are shown and described in more detail.

[0031] Figure 1 1 is a block diagram of a system 100 including a host device 105 and a memory device / data storage device 110 according to an example. In one example, the data storage device 110 can be equivalent to a memory device. For example, the memory device / data storage device 110 can be an MRAM device, a PCM device, a solid-state drive (SSD), or any other NAND-based memory device. In one example, the host device 105 includes a processor 115 and a memory 120 (e.g., a main memory). The memory 120 may include an operating system 125, a kernel 130, and / or an application 135 or otherwise be associated with the operating system, the kernel, and / or the application.

[0032] The processor 115 may execute various instructions, such as instructions from the operating system 125 and / or the application 135. The processor 115 may include circuits such as a microcontroller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), hardwired logic, analog circuits, and / or various combinations thereof. In one example, the processor 115 may include a system on a chip (SoC).

[0033] In one example, memory 120 may be used by host device 105 to store data for use or otherwise execution by processor 115. Data stored in memory 120 may include instructions provided by memory device / data storage device 110 via communication interface 140. Data stored in memory 120 may also include data used to execute instructions from operating system 125 and / or one or more application programs 135. Memory 120 may be a single memory or may include multiple memories, such as one or more non-volatile memories, one or more volatile memories, or a combination thereof.

[0034] In one example, operating system 125 may create a virtual address space for application 135 and / or other processes executed by processor 115. The virtual address space may be mapped to a location in memory 120. Operating system 125 may also include or otherwise be associated with kernel 130. Kernel 130 may include instructions for managing various resources of host device 105 (e.g., memory allocation), processing read and write requests, etc.

[0035] The communication interface 140 communicatively couples the host device 105 and the memory device / data storage device 110. The communication interface 140 may be a serial advanced technology attachment (SATA), a PCI express (PCIe) bus, a small computer system interface (SCSI), a serial attached SCSI (SAS), Ethernet, Fibre Channel, or Wi-Fi. Therefore, the host device 105 and the memory device / data storage device 110 do not need to be physically located at the same location, and can communicate via a network such as a local area network (LAN) or a wide area network (WAN) (such as the Internet). In addition, the host device 105 can be connected to the memory device / data storage device 110 using a logical interface specification such as a non-volatile memory standard (NVMe) or an advanced host controller interface (AHCI).

[0036] The memory device / data storage device 110 may include a controller 150 and a memory device 155. In one example, the controller 150 is communicatively coupled to the memory device 155. The memory device 155 may be an MRAM device, a PCM device, a solid state drive (SSD), or any other NAND-based memory device. Although specific examples are given, the memory device 155 may be any type of storage device to which data is written and from which data is read.

[0037] In one example, the memory device 155 includes one or more memory dies (e.g., a first memory die 165 and a second memory die 170). Although memory dies are specifically mentioned, the memory device 155 may include any non-volatile memory device, storage device, storage element, or storage medium, including NAND flash memory cells and / or NOR flash memory cells.

[0038] The memory cells may be in the form of solid-state (e.g., flash) memory cells and may be programmable once, programmable a few times, or programmable multiple times. Additionally, the memory cells may be single-level cells (SLC), multi-level cells (MLC), triple-level cells (TLC), quad-level cells (QLC), penta-level cells (PLCs), and / or use any other memory technology. The memory cells may be arranged in a two-dimensional configuration or a three-dimensional configuration.

[0039] In some examples, the memory device / data storage device 110 may be attached to or embedded within the host device 105. In another example, the memory device / data storage device 110 may be implemented as an external device or a portable device that is communicatively coupled to or selectively coupled to the host device 105. In yet another example, the memory device / data storage device 110 may be a component (e.g., a solid state drive (SSD)) of a network-accessible data storage system, a network-attached storage system, a cloud data storage system, etc.

[0040] The memory device 155 may also include support circuits. In one example, the support circuits include read / write circuits 160. The read / write circuits 160 support the operation of the memory die of the memory device 155. Although the read / write circuits 160 are depicted as a single component, the read / write circuits 160 may be divided into separate components, such as a read circuit and a write circuit. The read / write circuits 160 may be located outside the memory die of the memory device 155. In another example, one or more of the memory dies may include corresponding read / write circuits 160 that are operable to read data from and / or write data to storage elements within a separate memory die independently of other read and / or write operations on any other memory die.

[0041] In one example, one or more of the first memory die 165 and the second memory die 170 include one or more memory blocks. In one example, each memory block includes one or more memory cells. A block of memory cells is the smallest number of memory cells that can be physically erased together. In one example and to increase parallelism, each block in a block can be operated or organized in a larger block or metablock. For example, a block from different planes of memory cells can be logically linked together to form a metablock.

[0042] As previously described, the memory device / data storage device 110 may also include a controller 150. The controller 150 may be communicatively coupled to the memory device 155 via a bus, interface, or other communication circuit. In one example, the communication circuit may include one or more channels to enable the controller 150 to communicate with the first memory die 165 and / or the second memory die 170 of the memory device 155. In another example, the communication circuit may include multiple different channels to enable the controller 150 to communicate with the first memory die 165 independently and / or communicate with the second memory die 170 of the memory device 155 in parallel with the first memory die.

[0043] The controller 150 may receive data and / or instructions from the host device 105. The controller 150 may also send data to the host device 105. For example, the controller 150 may send data to and / or receive data from the host device 105 via the communication interface 140. The controller 150 may also send data and / or commands to and / or receive data from the memory device 155.

[0044] The controller 150 may send data and a corresponding write command to the memory device 155 to cause the memory device 155 to store the data at a specified address of the memory device 155. In one example, the write command specifies a physical address of a portion of the memory device 155. The controller 150 may also send data and / or commands associated with one or more background scan operations, garbage collection operations, and / or wear leveling operations.

[0045] The controller 150 may also send one or more read commands to the memory device 155. In one example, the read command specifies a physical address of a portion of the memory device 155 storing data. The controller 150 may also track the number of program / erase cycles or other programming operations that have been performed on or by the memory device and / or the memory die of the memory device 155.

[0046] The controller 150 may also include or otherwise be associated with a bypass buffer 180. In one example, the bypass buffer 180 stores codewords that have been identified as worn due to stuck-type fault bit errors and / or other persistent faults. For example, the controller 150 periodically checks the number of worn bits (e.g., bits of a codeword that are stuck at a logic "1" or logic "0" due to a stuck-type fault) in each codeword stored by the memory device 155. If the number of worn bits exceeds a threshold, the codeword is corrected (e.g., using an ECC operation) and stored as an entry in the bypass buffer 180. The memory address of the memory device 155 associated with the worn codeword is also stored in the bypass buffer 180 and associated with the corrected codeword.

[0047] When the controller 150 receives a subsequent read and / or write command, the controller 150 checks the bypass buffer 180 to determine whether the memory address associated with the received command is stored as an entry in the bypass buffer 180. If the memory address associated with the command is stored as an entry in the bypass buffer 180, the command is executed on the codeword stored in the bypass buffer 180. However, if the memory address associated with the command is not stored as an entry in the bypass buffer 180, the command is executed on the codeword stored in the memory address of the memory device 155.

[0048] The controller 150 may also include an error correction code (ECC) system 185 or otherwise be associated with an error correction code (ECC) system. In one example, the ECC system 185 may work with the bypass buffer 180. For example, the ECC system 185 may receive a codeword that has been identified as having a wear bit exceeding a threshold number. The ECC system 185 may correct errors and / or generate one or more ECC codewords based at least in part on the corrected data. For example, the ECC system 185 may include an encoder that encodes the received data using one or more encoding techniques. The ECC system 185 may also include a decoder for decoding the data read from the memory device 155. The ECC system 185 may also detect and correct bit errors that may exist in the stored / read data. For example, the ECC system 185 may correct multiple bit errors that are at most the error correction capability of the specific ECC technology used by the ECC system 185.

[0049] Figure 2 A bypass buffer 200 for storing worn codewords according to one example is shown. In one example, the bypass buffer 200 may be similar to that described with respect to Figure 1The bypass buffer 180 shown and described. The bypass buffer 200 may be configured as a content addressable memory (CAM) or some other associative memory that allows data to be retrieved based on the content of the data rather than based on a specific memory address.

[0050] In one example, the bypass buffer 200 includes n entries 210. For example, the bypass buffer 200 may include one thousand entries. Although one thousand entries are mentioned, the bypass buffer 200 may include any number of entries 210. Each entry 210 of the bypass buffer 200 may include a valid entry indicator 220, a memory address 230, and a codeword 240.

[0051] In one example, the codeword 240 of each entry 210 is a codeword identified as having more than a threshold number of worn bits. For example, if the controller determines that a codeword stored in a memory device (e.g., memory device 155 ( Figure 1 )) has x worn bits and further determines that x exceeds the worn bit threshold, the controller causes the error correction system (e.g., ECC system 185 ( Figure 1 )) corrects any errors in the codeword caused by worn bits, as well as any soft bit errors in the codeword. The controller may then cause the corrected codeword to be stored as entry 210 in bypass buffer 200.

[0052] The controller may also cause a memory address 230 of the memory device from which the original codeword was read to be stored in the bypass buffer 200. In one example, the memory address 230 is associated with the corrected codeword 240 in the bypass buffer 200.

[0053] The bypass buffer 200 also includes a valid entry indicator 220. In one example, the valid entry indicator 220 is a bit or other indicator that informs the controller that a particular entry 210 in the bypass buffer 200 includes valid data (e.g., a corrected codeword 240 and a memory address 230 associated with the original read codeword) rather than data used to initialize the bypass buffer 200.

[0054] Additional information about bypass buffer 200 and how to fill and use bypass buffer 200 will be referred to Figures 3 to 7 Explain in more detail.

[0055] Figure 3 A method 300 for determining whether a codeword should be added to a bypass buffer according to an example is shown. In one example, the method 300 may be performed by a controller, such as Figure 1 Executed or otherwise performed by the controller 150 as shown and described. Figure 3 The bypass buffer described may be about Figure 1 The bypass buffer 180 shown and described and / or with respect to Figure 2 Bypass buffer 200 is shown and described.

[0056] Method 300 performs the following steps in which a memory device (eg, memory device 155 ( Figure 1 )) begins when a codeword is read (310) from a memory address of the memory device. In one example, the controller may periodically read some or all of the codewords stored by the memory device to determine whether one or more codewords should be added to the bypass buffer.

[0057] In another example, the controller may read one or more codewords from the memory device based on one or more trigger conditions. In one example, the trigger conditions may include, but are not limited to, the age of the memory device, the number of program / erase cycles performed by the memory device, the amount of time / resources spent performing ECC operations, etc. In another example, a codeword associated with a particular address range in the memory device may be selected (e.g., manually or automatically) and subjected to the method 300 described herein.

[0058] Once the codeword is read from the memory address of the memory device, the raw codeword data is stored (320) in a temporary memory location. In one example, the temporary memory location may be in a bypass buffer. In another example, the temporary memory location may be a reserved memory location in the memory device. As part of this process, the codeword may undergo an error correction code process to help ensure that the data associated with the codeword is accurate / correct.

[0059] When the codeword has been stored in the temporary memory location, the controller writes (330) test data to the memory address of the memory device from which the codeword was read. In one example, the test data includes a plurality of bits and each bit has the same logical value. For example, the test data may include all 1 bits or all 0 bits. In one example, the logical value of each bit of the test data may depend on how the stuck-at fault manifests.

[0060] For example, if a stuck-at fault of a codeword is represented as a 1 bit, the test data includes all 0 bits. Similarly, if a stuck-at fault manifests as a 0 bit, the test data includes all 1 bits. Figure 3 In the example of , a stuck-at fault manifests itself as a 1 bit. Thus, the test data includes all 0 bits. However, as previously explained, the test data may have bits in a first state, while a stuck-at fault bit may be represented by a bit having a second, different state.

[0061] Once the test data has been written to the memory address of the memory device, the controller determines 340 the number of 1 bits in the codeword. Since the test data includes all 0 bits, the number of 1 bits read from the memory address indicates the number of worn bits of the codeword.

[0062] The controller may then restore (350) the original data (or corrected data) associated with the codeword to the memory address of the memory device from which the codeword was read (e.g., in operation 310). In one example, restoring the original data associated with the codeword includes reading the codeword from the temporary memory location and writing the codeword to the memory address of the memory device.

[0063] The controller may also identify (360) or otherwise determine the number of worn bits of the codeword. As previously indicated, the number of worn bits of the codeword is based at least in part on the number of bits having logic states opposite or different than the logic states of the bits of the codeword that include the test data.

[0064] The controller may then determine (370) whether the number of worn bits for the codeword exceeds a worn bit threshold. If it is determined that the number of worn bits for the codeword does not exceed the worn bit threshold, method 300 may be repeated for another codeword.

[0065] However, if it is determined (370) that the number of worn bits of the codeword exceeds the worn bit threshold, the controller may also determine (380) whether the bypass buffer already includes an entry for the codeword. As part of this process, the controller may determine whether the bypass buffer includes an entry that includes the codeword and an entry for a memory location of the memory device where the codeword was originally stored. To make this determination, the controller may compare a memory address associated with the memory location from which the codeword was read to each memory address field (e.g., memory address 230( Figure 2 )) for comparison.

[0066] If the controller determines (380) that the bypass buffer already includes an entry for the particular codeword / memory address, the method 300 may be repeated for another codeword. However, if the controller determines (380) that the bypass buffer does not include an entry for the particular codeword / memory address, the controller may update (390) the bypass buffer.

[0067] In one example, updating the bypass buffer includes performing an ECC operation on the codeword (if necessary). The data associated with the codeword is then stored in the codeword field (e.g., codeword 240 ( Figure 2 The memory address of the memory device from which the codeword was originally read is also stored in the bypass buffer (e.g., as memory address 230 ( Figure 2)) and is associated with the codeword. A valid entry indicator (eg, valid entry indicator 220 ( Figure 2 )), thereby indicating that the particular entry in the bypass buffer has been updated with the corrected codeword. Method 300 may then be repeated for another codeword.

[0068] In one example, the controller may also determine whether the bypass buffer has room for additional entries before updating the bypass buffer. For example, and as previously described, the bypass buffer may have a limited number of (e.g., one thousand) entries. If all entries are taken, then additional codewords may not be added to the bypass buffer.

[0069] Figure 4 A method 400 for performing a read operation on a bypass buffer according to one example is shown. In one example, the method 400 may be performed by a controller (e.g., Figure 1 Likewise, the bypass buffer may be similar to that described with respect to Figure 1 The bypass buffer 180 shown and described and / or with respect to Figure 2 Bypass buffer 200 is shown and described.

[0070] The method 400 begins when a controller receives (410) a read operation. In one example, the read operation is received from a host device. When the read operation is received, the controller determines (420) a memory address associated with the read operation.

[0071] The controller may then compare the memory address associated with the read operation to various memory address entries in the bypass buffer (eg, memory address 230 ( Figure 2 )) is compared (430). If the controller determines (440) that a matching memory address is found as an entry in the bypass buffer, the controller returns (450) the codeword associated with the memory address (e.g., codeword 240 ( Figure 2 )). However, if the controller determines (440) that a matching entry is not found in the bypass buffer, the controller retrieves the matching entry from a memory device (eg, memory device 155 ( Figure 1 )) reads the codeword from the memory address, performs any error correction code operations (if necessary), and returns (470) the codeword from the memory device.

[0072] Figure 5 Another method 500 for performing a read operation on a bypass buffer according to one example is shown. Figure 5 The examples shown and described are similar to those described with respect to Figure 4However, in this example, the read operation performed on the bypass buffer can be performed in parallel or substantially in parallel with the read operation performed on the memory address of the memory device. Because the read operation is performed in parallel or substantially in parallel (with respect to Figure 4 Compared to the sequential read shown and described), the performance of the memory device can be improved.

[0073] Additionally, in the following examples, the method 500 may be performed by a controller (eg, Figure 1 In addition, the method 500 may be performed in a manner similar to that described with respect to Figure 1 The bypass buffer 180 and memory device 155 shown and described and / or with respect to Figure 2 The bypass buffer 200 shown and described is executed on a bypass buffer and / or a memory device.

[0074] Method 500 begins when a controller receives (510) a read operation. In one example, from a host device (e.g., host device 105 ( Figure 1 ))Receive a read operation. When the read operation is received, the controller determines (520) a memory address associated with the read operation.

[0075] The controller may then compare the memory address associated with the read operation to the memory address of each entry in the bypass buffer (eg, memory address 230 ( Figure 2 )) for comparison (530). Additionally, and in parallel or substantially in parallel, the controller reads (540) a codeword from the memory address of the memory device. The controller may then determine (540) whether a matching memory address is found in the bypass buffer. If a matching memory address is found in the bypass buffer, the controller returns (560) the codeword associated with the memory address from the bypass buffer. The codeword read from the memory address of the memory device may be discarded.

[0076] However, if the controller determines (540) that no matching entry is found in the bypass buffer, the controller may perform ECC operations (if necessary) on the codeword read from the memory device (eg, from operation 540) and return (570) the codeword read from the memory device.

[0077] Figure 6 A method 600 for performing a write operation on a bypass buffer according to one example is shown. In one example, the method 600 can be performed by a controller such as described herein. Likewise, the bypass buffer and / or memory device can be similar to the various bypass buffers and memory devices described herein.

[0078] The method 600 begins when a controller receives (610) a write operation. In one example, the write operation is received from a host device. When the write operation is received, the controller determines (620) a memory address associated with the write operation.

[0079] The controller may then compare (630) the memory address associated with the write operation to each entry in the bypass buffer to determine whether the bypass buffer includes the memory address associated with the write operation. If the controller determines (640) that a matching memory address is found in the bypass buffer, the controller writes (650) the data associated with the write operation to the bypass buffer. For example, the data associated with the write operation is written to the codeword field (e.g., codeword 240 ( Figure 2 ))middle.

[0080] However, if the controller determines (640) that no matching entry is found in the bypass buffer (eg, the memory address associated with the write command is not in the bypass buffer), the controller writes (660) the data associated with the write command to the memory address of the memory device.

[0081] Figure 7 Another method 700 for performing a write operation on a bypass buffer according to one example is shown. Figure 7 The examples shown and described are similar to those described with respect to Figure 6 However, in this example, the write operation performed on the bypass buffer can be performed in parallel or substantially in parallel with the write operation performed on the memory device. Because the write operation is performed in parallel or substantially in parallel (with respect to Figure 6 Compared with the sequential writing shown and described), the performance of the memory device can be improved.

[0082] Additionally, in the following examples, the method 700 may be performed by a controller such as previously described herein. Likewise, the bypass buffer and / or memory device may be similar to the various bypass buffers and memory devices described herein.

[0083] Method 700 begins when a controller receives (710) a write operation. In one example, from a host device (e.g., host device 105 ( Figure 1 )) receives a write operation. When receiving the write operation, the controller determines (720) a memory address associated with the write operation. For example, the controller determines (720) a memory address of a memory device to which data associated with the write command is to be written.

[0084] The controller may then compare the memory address associated with the write operation to various memory addresses in the bypass buffer (eg, memory address 230 ( Figure 2 )) entries are compared (730). Additionally, and in parallel or substantially in parallel, the controller writes (740) data associated with the write command to a specified memory address of the memory device.

[0085] The controller may also determine (750) whether a matching address is found in the bypass buffer. If a matching memory address is found in the bypass buffer, the controller writes (760) (e.g., overwrites) the codeword stored in the bypass buffer that is associated with the memory address of the newly received data. However, if the controller determines (750) that a matching entry is not found in the bypass buffer, the method ends (770) because the data has been written to the correct memory address of the memory device.

[0086] Figure 8 8 shows a bypass buffer 800 with a confidence indicator 850 according to one example. The bypass buffer 800 may be similar to the Figure 1 Bypass buffer 180 is shown and described. In one example, bypass buffer 800 may be configured as a content addressable memory (CAM) or some other associated memory such as previously described.

[0087] In one example, the bypass buffer 800 includes n entries 810. For example, the bypass buffer 800 may include one thousand entries. Although one thousand entries are mentioned, the bypass buffer 800 may include any number of entries. Each entry 810 of the bypass buffer 800 may include a valid entry indicator 820, a memory address 830, a codeword 840 (or data associated with the codeword), and a confidence indicator 850.

[0088] In one example, the codeword 840 of each entry 810 is a codeword identified as having more than a threshold number of worn bits. For example, if the controller determines that the codeword 840 stored in the memory device (e.g., memory device 155 ( Figure 1 )) has x worn bits, the controller causes the error correction system (e.g., ECC system 185 ( Figure 1 The controller may also cause the corrected codeword to be stored as entry 810 in bypass buffer 800.

[0089] The controller may also store a memory address 830 of the memory device from which the original codeword was read in the bypass buffer 800. In one example, the memory address 830 is associated with the corrected codeword 840 in the bypass buffer 800.

[0090] The bypass buffer 800 also includes a valid entry indicator 820. In one example, the valid entry indicator 820 is a bit or other indicator that informs the controller that a particular entry 810 in the bypass buffer includes valid data (e.g., a corrected codeword 840 and a memory address associated with the original read codeword) rather than data used to initialize the bypass buffer.

[0091] The bypass buffer 800 may also include a confidence indicator 850. The confidence indicator 850 may be used to determine whether any codewords and / or associated memory addresses have been erroneously added (e.g., false positives) to the bypass buffer 800. For example, during the determination of whether the number of worn bits of a codeword exceeds a worn bit threshold (e.g., operation 370 ( Figure 3 )), the number of worn bits of a codeword may be incorrectly determined or otherwise incorrectly reported.

[0092] Thus, when a matching entry (e.g., codeword 840 and / or memory address) is found in the bypass buffer, the confidence indicator 850 can be incremented, and when a matching entry is not found in the bypass buffer, the confidence indicator can be decremented. In one example, the confidence indicator 850 can be incremented until saturated (e.g., until a matching entry is found in the bypass buffer y times). In another example, if the confidence indicator is decremented (e.g., because a matching entry is not found in the bypass buffer) and decrements to zero, the entry in the bypass buffer is removed (e.g., flipping a bit associated with the valid entry indicator 820).

[0093] Additional information about bypass buffer 800 and how to fill and use bypass buffer 800 is provided in Fig. 9 Explain in more detail.

[0094] Fig. 9 A method 900 for determining whether a codeword should be added to a bypass buffer according to an example is shown. In one example, the method 900 can be performed by a controller such as previously described herein. Similarly, the method 900 can be used or combined with the method of Figure 8 The bypass buffer 800 shown and described is used.

[0095] In this example, operations of reading a codeword, storing the original codeword data, and writing test data to the memory address associated with the read command have been performed. These operations may be similar to those described with respect to Figure 3Operations 310 to 330 are shown and described. Once these operations have been performed, the controller can determine (905) the number of 1 bits in the codeword. As previously discussed, although the number of 1 bits is specifically described, once the test data has been written to the memory address associated with the received read command, the controller can determine the number of 0 bits in the codeword.

[0096] The controller may then determine (910) whether the number of worn bits in the codeword exceeds a worn bit threshold. If it is determined (910) that the number of worn bits in the codeword exceeds the worn bit threshold, the controller may further determine (915) whether there is a valid entry in the bypass buffer. In one example, the controller may determine, based at least in part on comparing a memory address associated with the memory location from which the codeword was read with each memory address field (e.g., memory address 830 ( Figure 8 )) to make this determination.

[0097] If the controller determines (915) that the bypass buffer already includes an entry for the particular codeword / memory address, the confidence indicator associated with the particular entry is incremented (920).The method 900 may then be repeated for another codeword.

[0098] However, if the controller determines (915) that the bypass buffer does not include an entry for the particular codeword / memory address, the controller may update (925) the bypass buffer.

[0099] In one example, updating the bypass buffer includes performing an error correction code operation on the codeword (if necessary). Data associated with the codeword may then be stored in a codeword field (e.g., codeword 840 ( Figure 8 )). The memory address of the memory device from which the codeword was originally read is also stored in the bypass buffer (e.g., as memory address 830) and associated with the codeword. A valid entry indicator (e.g., valid entry indicator 820 ( Figure 8 )), thereby indicating that the particular entry in the bypass buffer has been updated with the corrected codeword. In addition, the confidence indicator associated with the entry is set (e.g., set to 1) to indicate that the entry is a newly added entry. Method 900 can then be repeated for another codeword.

[0100] In one example, the controller may also determine whether the bypass buffer has room for additional entries before updating the bypass buffer. For example, and as previously described, the bypass buffer may have a limited number of entries. Therefore, if all entries are taken, additional codewords may not be added to the bypass buffer.

[0101] Referring again to operation 910, if the controller determines that the worn bits of the particular codeword do not exceed the worn bit threshold, the controller may then determine (930) whether there is a valid entry associated with the codeword in the bypass buffer. For example, the controller may determine whether the particular codeword was erroneously added to the bypass buffer. In one example, the determination is based at least in part on comparing a memory address associated with the memory location from which the codeword was read with each memory address field (e.g., memory address 830 ( Figure 8 )) for comparison.

[0102] If it is determined (930) that a valid entry associated with the codeword is not in the bypass buffer, then method 900 may end (935).Method 900 may then be repeated for another codeword.

[0103] However, if it is determined (930) that a valid entry associated with the codeword is in the bypass buffer (e.g., the codeword has (possibly) been erroneously added to the bypass buffer), the controller decrements (940) the confidence indicator associated with the entry. Once the confidence indicator has been decremented, the controller may determine (945) whether the confidence indicator is zero. If the controller determines that the confidence indicator is not zero (or greater than zero), the method ends (935). The method 900 may then be repeated for another codeword.

[0104] However, if it is determined that the confidence indicator associated with the entry is equal to zero (or less than zero), then the bypass buffer entry associated with the particular codeword and / or memory address is deallocated. In one example, the bypass buffer entry associated with the particular codeword and / or memory address is deallocated by switching / flipping a bit associated with the valid entry indicator.

[0105] Figure 10 to Figure 11 Exemplary storage devices that can be used with or otherwise implement the various features described herein are described. Figure 10 to Figure 11 The storage devices shown and described may include Figure 1 Various systems and components similar to those shown and described. Fig.11 The controller 1122 shown and described may be similar to Figure 1 Likewise, the memory die 1108 may be similar to Figure 1 The first memory die 165 and / or the second memory die 170.

[0106] Fig.101 is a perspective view of a memory device 1000 including a three-dimensional (3D) stacked non-volatile memory according to one example. In this example, the memory device 1000 includes a substrate 1010. Blocks of memory cells are included on or above the substrate 1010. The blocks may include a first block (BLK0 1020) and a second block (BLK1 1030). Each block may be formed of a memory cell (e.g., a non-volatile memory element). The substrate 1010 may also include a peripheral region 1040 having support circuits used by the first block and the second block.

[0107] The substrate 1010 may also carry circuits below the blocks and one or more lower metal layers patterned in the conductive paths to carry signals from the circuits. The blocks may be formed in the middle region 1050 of the memory device 1000. The memory device may also include an upper region 1060. The upper region 1060 may include one or more upper metal layers that are patterned in the conductive paths to carry signals from the circuits. Each block of memory cells may include a stacking region of memory cells. In one example, the alternating levels of the stack represent word lines. Although two blocks are depicted, additional blocks may be used and the additional blocks extend in the x-direction and / or the y-direction.

[0108] In one example, the length of the plane of the substrate 1010 in the x direction represents the direction in which a signal path of a word line or a control gate line extends (e.g., a word line or a drain terminal select gate (SGD) line direction), and the width of the plane of the substrate 1010 in the y direction represents the direction in which a signal path of a bit line extends (e.g., a bit line direction). The z direction represents the height of the memory device 1000.

[0109] Fig.11 is a functional block diagram of a storage device 1100 according to an example. In one example, the storage device 1100 may be Fig.10 A 3D stacked non-volatile memory device 1000 is shown and described. Fig.11The components depicted in can be circuits. In one example, the storage device 1100 includes one or more memory dies 1105. Each memory die 1105 includes a three-dimensional memory structure 1110 of memory cells (e.g., a 3D array of memory cells), a control circuit 1115, and a read / write circuit 1120. In another example, a two-dimensional array of memory cells can be used. The memory structure 1110 can be addressed by word lines using a first decoder 1125 (e.g., a row decoder) and by bit lines using a second decoder 1130 (e.g., a column decoder). The read / write circuit 1120 may also include a plurality of sense blocks 1135, including SB1, SB2...SBp (e.g., a sense circuit), which allow pages of memory cells to be read or programmed in parallel. The sense block 1135 may include a bit line driver.

[0110] In one example, the controller 1140 is included in the same memory device 1100 as one or more memory dies 1105. In another example, the controller 1140 is formed on a die bonded to the memory dies 1105, in which case each memory die 1105 may have its own controller 1140. In yet another example, the controller die controls all of the memory dies 1105.

[0111] Commands and data may be transferred between the host 1145 and the controller 1140 using a data bus 1150. Commands and data may also be transferred between the controller 1140 and one or more memory dies 1105 via lines 1155. In one example, the memory die 1105 includes a set of input and / or output (I / O) pins connected to the lines 1155.

[0112] The memory structure 1110 may also include one or more memory cell arrays. The memory cells may be arranged in a three-dimensional array or a two-dimensional array. The memory structure 1110 may include any type of non-volatile memory formed on an array of memory cells having one or more physical levels of active regions disposed above a silicon substrate. The memory structure 1110 may be in a non-volatile memory device having circuitry associated with the operation of the memory cells, whether the associated circuitry is above or within the substrate.

[0113] Control circuitry 1115 works in conjunction with read / write circuitry 1120 to perform memory operations (eg, erase, program, read, etc.) on memory structure 1110. Control circuitry 1115 may include registers, ROM fuses, and other devices for storing default values ​​such as reference voltages and other parameters.

[0114] The control circuit 1115 may also include a state machine 1160, an on-chip address decoder 1165, a power control module 1170, and a temperature detection circuit 1175. The state machine 1160 may provide chip-level control of various memory operations. The state machine 1160 may be programmable by software. In another example, the state machine 1160 does not use software and is implemented entirely in hardware (e.g., electronic circuits).

[0115] The on-chip address decoder 1165 may provide an address interface between addresses used by the host 1145 and / or the controller 1140 and hardware addresses used by the first decoder 1125 and the second decoder 1130 .

[0116] The power control module 1170 may control the power and voltage supplied to the word lines and bit lines during memory operations. The power control module 1170 may include drivers for the word line layers in a 3D configuration, select transistors (e.g., SGS and SGD transistors), and source lines. The power control module 1170 may include one or more charge pumps for generating voltages.

[0117] The control circuit 1115 may also include a temperature detection circuit 1175. The temperature detection circuit may be configured to detect the temperature of one or more components of the memory device 1100.

[0118] The control circuit 1115, state machine 1160, on-chip address decoder 1165, first decoder 1125, second decoder 1130, temperature detection circuit 1175, power control module 1170, sensing block 1135, read / write circuit 1120 and / or controller 1140 can be considered as one or more control circuits and / or management circuits that perform some or all of the operations described herein.

[0119] In one example, the controller 1140 is a circuit that can be on-chip or off-chip. In addition, the controller 1140 may include one or more processors 1180, ROM 1185, RAM 1190, memory interface 1195, and host interface 1197, all of which can be interconnected. In one example, one or more processors 1180 are an example of a control circuit. Other examples may use a state machine or other custom circuit designed to perform one or more functions. Devices such as ROM 1185 and RAM 1190 may include code, such as an instruction set. One or more processors in the processor 1180 are operable to execute the set of instructions to provide some or all of the functions described herein.

[0120] Alternatively or in addition, one or more of the processors 1180 may access code from a memory device in the memory structure 1110, such as a reserved area of ​​memory cells connected to one or more word lines. A memory interface 1195 in communication with the ROM 1185, the RAM 1190, and the one or more processors 1180 may be a circuit that provides an electrical interface between the controller 1140 and the memory die 1105. For example, the memory interface 1195 may change the format or timing of a signal, provide a buffer, isolate surges, latch I / O, etc.

[0121] The one or more processors 1180 may use the memory interface 1195 to issue commands to the control circuit 1115 or any other component of the memory die 1105. The host interface 1197, which communicates with the ROM 1185, the RAM 1195, and the one or more processors 1180, may be a circuit that provides an electrical interface between the controller 1140 and the host 1145. For example, the host interface 1197 may change the format or timing of signals, provide buffers, isolate surges, latch I / O, etc. Commands and data from the host 1145 are received by the controller 1140 through the host interface 1197. Data sent to the host 1145 may be transmitted using the data bus 1150.

[0122] The multiple memory elements in the memory structure 1110 can be configured so that they are connected in series or so that each element is individually accessible. As a non-limiting example, a flash memory device in a NAND configuration (e.g., NAND flash memory) typically contains memory elements connected in series. A NAND string is an example of a group of memory cells and select gate transistors connected in series.

[0123] A NAND flash memory array can also be configured so that the array includes multiple NAND strings. In one example, a NAND string includes multiple memory cells that share a single bit line and are accessed as a group. Alternatively, the memory elements can be configured so that each memory element can be accessed individually (e.g., a NOR memory array). NAND and NOR memory configurations are examples and the memory cells can have other configurations.

[0124] Memory cells may be arranged in an ordered array in a single memory device level, such as in multiple rows and / or columns. However, memory elements may be arranged in irregular or non-orthogonal configurations, or in structures that are not considered arrays.

[0125] In one example, a 3D memory structure may be arranged vertically as a stack of multiple 2D memory device levels. As another non-limiting example, a 3D memory array may be arranged as a plurality of vertical columns (e.g., columns substantially perpendicular to the major surface of the substrate such as extending in the y-direction), wherein each column has a plurality of memory cells. The vertical columns may be arranged into a two-dimensional arrangement of memory cells, wherein the memory cells are located on a plurality of vertically stacked memory planes. Other configurations of three-dimensional memory elements may also constitute a 3D memory array.

[0126] In another example, in a 3D NAND memory array, memory elements can be coupled together to form vertical NAND strings that span multiple horizontal memory device levels. Other 3D configurations can be envisioned, where some NAND strings contain memory elements in a single memory level, while other strings contain memory elements that span multiple memory levels. The 3D memory array can also be designed in a NOR configuration and in a ReRAM configuration.

[0127] Those of ordinary skill in the art will recognize that the technology described herein is not limited to a single specific memory structure, but encompasses many related memory structures within the spirit and scope of the technology described herein and as understood by those of ordinary skill in the art.

[0128] The term computer-readable media as used herein may include computer storage media. Computer storage media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, or program modules. Computer storage media may include RAM, ROM, electrically erasable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage devices, magnetic cassettes, magnetic tape, magnetic disk storage devices or other magnetic storage devices, or may be used to store information and may be used by a computing device (e.g., host device 105 ( Figure 1 )) or any other article of manufacture accessible by the computer. Any such computer storage medium may be part of a computing device. Computer storage media does not include carrier waves or other propagated or modulated data signals.

[0129] Additionally, the examples described herein may be discussed in the general context of computer-executable instructions residing on some form of computer-readable storage media, such as program modules executed by one or more computers or other devices. By way of example and not limitation, computer-readable storage media may include non-transitory computer storage media and communication media. In general, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The functionality of the program modules may be combined or distributed in the various examples as desired.

[0130] Communication media may be embodied by computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and includes any information delivery media. The term "modulated data signal" may describe a signal that has one or more characteristics set or changed in a manner that encodes information in the signal. By way of example and not limitation, communication media may include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0131] Thus, an example of the present disclosure describes a method, the method comprising: reading data associated with a codeword located at a memory address of a memory device; storing the data associated with the codeword in a temporary storage location; writing test data to the memory address associated with the codeword, the test data comprising a plurality of bits, wherein each bit is in a first state; reading the test data from the memory address associated with the codeword; determining the number of bits in the test data having a second state different from the first state; and based at least in part on determining that the number of bits in the second state is above a threshold: storing the data associated with the codeword as an entry in a bypass buffer; and associating the entry in the bypass buffer with the memory address of the memory device. In one example, the method further comprises updating a valid entry indicator associated with the entry in the bypass buffer based at least in part on the data associated with the codeword being stored as an entry in the bypass buffer. In one example, the method further comprises writing the data associated with the codeword from the temporary storage location to the memory address of the memory device based at least in part on determining that the number of bits in the second state is below a threshold. In one example, the method further comprises performing an error correction code operation on the data associated with the codeword based at least in part on the data associated with the codeword being read from the memory address of the memory device. In one example, the method further includes receiving a read request associated with a codeword located at a specific memory address of a memory device; determining whether the specific memory address is associated with a specific entry in a bypass buffer; and returning data associated with the specific entry based at least in part on determining that the specific memory address is associated with the specific entry in the bypass buffer. In one example, the method further includes receiving a read request associated with a codeword located at a specific memory address of a memory device; reading data associated with the codeword from the specific memory address of the memory device; determining whether the specific memory address is associated with a specific entry in a bypass buffer; and returning data associated with the specific entry based at least in part on determining that the specific memory address is associated with the specific entry in the bypass buffer. In one example, the method further includes returning data read from the specific memory address of the memory device based at least in part on determining that the specific memory address is not present in the bypass buffer. In one example, the method further includes receiving a write request associated with a specific memory address of a memory device; determining whether the specific memory address is associated with a specific entry in a bypass buffer; and writing data associated with the write request to the specific entry in the bypass buffer based at least in part on determining that the specific memory address is associated with the specific entry in the bypass buffer. In one example, the method further includes writing data associated with the write request to a particular memory address of the memory device based at least in part on determining that the particular memory address does not exist in the bypass buffer.In one example, data associated with the write request is written to a specific memory address of the memory device, while data associated with the write request is written to a specific entry in the bypass buffer. In one example, the method also includes updating a confidence indicator associated with the entry in the bypass buffer based at least in part on storing the data associated with the codeword as an entry in the bypass buffer.

[0132] In another example, the present disclosure describes a data storage device, the data storage device comprising: a controller; and a memory, the memory being communicatively coupled to the controller and storing instructions, which when executed by the controller, cause the controller to perform the following operations: read data associated with a codeword located at a memory address of a memory device of the data storage device; store the data associated with the codeword in a temporary storage location; write test data to the memory address to determine the number of worn bits associated with the codeword, the test data comprising a plurality of bits, each of which is in the same state; read the test data from the memory address; determine the number of bits in different states in the test data; and based at least in part on determining that the number of bits in different states is above a threshold: store the data associated with the codeword as an entry in a bypass buffer; and associate the entry in the bypass buffer with the memory address of the memory device. In one example, the instructions also include instructions for updating a valid entry indicator associated with the entry in the bypass buffer based at least in part on the data associated with the codeword being stored as an entry in the bypass buffer. In one example, the instructions also include instructions for writing the data associated with the codeword from the temporary storage location to the memory address of the memory device based at least in part on determining that the number of bits in different states is below a threshold. In one example, the instructions also include instructions for performing an error correction code operation on data associated with the codeword based at least in part on the data associated with the codeword being read from a memory address of a memory device. In one example, the instructions also include instructions for performing the following operations: receiving a read request associated with a codeword located at a specific memory address of a memory device; determining whether the specific memory address is associated with a specific entry in a bypass buffer; and returning data associated with the specific entry based at least in part on determining that the specific memory address is associated with a specific entry in the bypass buffer. In one example, the instructions also include instructions for performing the following operations: receiving a read request associated with a codeword located at a specific memory address of a memory device; reading data associated with the codeword from a specific memory address of a memory device; determining whether the specific memory address is associated with a specific entry in a bypass buffer; and returning data associated with the specific entry based at least in part on determining that the specific memory address is associated with a specific entry in the bypass buffer. In one example, the instructions also include instructions for: receiving a write request associated with a specific memory address of a memory device; determining whether the specific memory address is associated with a specific entry in a bypass buffer; and writing data associated with the write request to the specific entry in the bypass buffer based at least in part on determining that the specific memory address is associated with the specific entry in the bypass buffer.

[0133] In another example, the present disclosure describes a data storage device, the data storage device comprising: means for reading data associated with a codeword located at a memory address of a memory device of the data storage device; means for storing the data associated with the codeword in a temporary storage location; means for writing test data to the memory address; means for determining a number of worn bits associated with the codeword, the number of worn bits being based at least in part on a number of bits having a first state associated with the test data and a number of bits having a second state different from the first state; means for storing the data associated with the codeword as an entry in a bypass buffer based at least in part on determining that the number of bits in the second state is above a threshold; and means for associating the entry in the bypass buffer with the memory address of the memory device. In one example, the data storage device also includes means for updating a valid entry indicator associated with the entry in the bypass buffer.

[0134] The description and illustration of one or more aspects provided in this disclosure are not intended to limit or restrict the scope of this disclosure in any way. The aspects, examples, and details provided in this disclosure are considered sufficient to convey ownership and enable others to make and use the best mode of the disclosure claimed.

[0135] Whether shown and described in combination or individually, various features (both structural and method features) are intended to be selectively rearranged, included, or omitted to produce an embodiment with a particular set of features. Having provided the description and illustration of the present application, those skilled in the art may envision changes, modifications, and alternative aspects that fall within the spirit of the broader aspects of the overall inventive concept embodied in the present application without departing from the broader scope of the disclosure claimed.

[0136] Various aspects of the present disclosure have been described below with reference to the schematic flow charts and / or schematic block diagrams of the methods, devices, systems, and computer program products according to the embodiments of the present disclosure. It should be understood that each frame of the schematic flow chart and / or schematic block diagram, and the combination of frames in the schematic flow chart and / or schematic block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor or other programmable data processing device of a computer to produce a machine, so that the instructions executed by the processor or other programmable data processing device create a device for implementing the functions and / or actions specified in one or more frames of the schematic flow chart and / or schematic block diagram.

[0137] References to elements using names such as "first," "second," etc. herein generally do not limit the number or order of those elements. Instead, these names can be used as a convenient method to distinguish between two or more elements or instances of elements. Thus, reference to a first element and a second element does not mean that only two elements can be used, or that the first element precedes the second element. Additionally, unless otherwise stated, a group of elements may include one or more elements.

[0138] A term of the form "at least one of A, B, or C" or "A, B, C, or any combination thereof" used in the specification or claims means "A or B or C, or any combination of these elements." For example, this term may include A, or B, or C, or A and B, or A and C, or A and B and C, or 2A, or 2B, or 2C, or 2A and B, etc. As an additional example, "at least one of A, B, or C" is intended to cover A, B, C, AB, AC, BC, and ABC, and multiples of the same member. Similarly, "at least one of A, B, and C" is intended to cover A, B, C, AB, AC, BC, and ABC, and multiples of the same member.

[0139] Similarly, as used herein, a phrase referring to a list of items linked with "and / or" refers to any combination of items. As an example, "A and / or B" is intended to cover A alone, B alone, or A and B together. As another example, "A, B, and / or C" is intended to cover A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B, and C together.

Claims

1. A method comprising: reading data associated with a codeword located at a memory address of a memory device; storing the data associated with the codeword in a temporary storage location; writing test data to the memory address associated with the codeword, the test data comprising a plurality of bits, wherein each bit is in a first state; reading the test data from the memory address associated with the codeword; determining a number of bits in the test data having a second state different from the first state; and Based at least in part on determining that the number of the bits in the second state is above a threshold: storing the data associated with the codeword as an entry in a bypass buffer; as well as The entry in the bypass buffer is associated with the memory address of the memory device.

2. The method of claim 1 , further comprising updating a valid entry indicator associated with the entry in the bypass buffer based at least in part on the data associated with the codeword being stored as the entry in the bypass buffer.

3. The method of claim 1 , further comprising writing the data associated with the codeword from the temporary storage location to the memory address of the memory device based at least in part on determining that the number of the bits in the second state is below the threshold. 4 . The method of claim 1 , further comprising performing an error correction code operation on the data associated with the codeword based at least in part on the data associated with the codeword being read from the memory address of the memory device.

5. The method according to claim 1, further comprising: receiving a read request associated with a codeword located at a particular memory address of the memory device; determining whether the particular memory address is associated with a particular entry in the look-aside buffer; as well as Based at least in part on determining that the particular memory address is associated with the particular entry in the bypass buffer, data associated with the particular entry is returned.

6. The method of claim 1 , further comprising receiving a read request associated with a codeword located at a particular memory address of the memory device; reading data associated with the codeword from the particular memory address of the memory device; determining whether the particular memory address is associated with a particular entry in the look-aside buffer; as well as Based at least in part on determining that the particular memory address is associated with the particular entry in the bypass buffer, data associated with the particular entry is returned.

7. The method of claim 6, further comprising returning the data read from the particular memory address of the memory device based at least in part on determining that the particular memory address does not exist in the bypass buffer.

8. The method of claim 1 , further comprising receiving a write request, the write request being associated with a particular memory address of the memory device; determining whether the particular memory address is associated with a particular entry in the look-aside buffer; as well as Based at least in part on determining that the particular memory address is associated with the particular entry in the bypass buffer, data associated with the write request is written to the particular entry in the bypass buffer.

9. The method of claim 8, further comprising writing the data associated with the write request to the particular memory address of the memory device based at least in part on determining that the particular memory address does not exist in the bypass buffer.

10. The method of claim 9, wherein the data associated with the write request is written to the specific memory address of the memory device simultaneously with the data associated with the write request being written to the specific entry in the bypass buffer.

11. The method of claim 1 , further comprising updating a confidence indicator associated with the entry in a bypass buffer based at least in part on storing the data associated with the codeword as the entry in the bypass buffer.

12. A data storage device comprising: Controller; and a memory communicatively coupled to the controller and storing instructions that, when executed by the controller, cause the controller to: reading data associated with a codeword located at a memory address of a memory device of the data storage device; storing the data associated with the codeword in a temporary storage location; writing test data to the memory address to determine a number of worn bits associated with the codeword, the test data comprising a plurality of bits wherein each bit is in a same state; Reading the test data from the memory address; determining the number of bits in the test data having different states; and Based at least in part on determining that the number of the bits in the different states is above a threshold: storing the data associated with the codeword as an entry in a bypass buffer; as well as The entry in the bypass buffer is associated with the memory address of the memory device.

13. The system of claim 12, further comprising instructions for updating a valid entry indicator associated with the entry in the bypass buffer based at least in part on the data associated with the codeword being stored as the entry in the bypass buffer.

14. The system of claim 12, further comprising instructions for writing the data associated with the codeword from the temporary storage location to the memory address of the memory device based at least in part on determining that the number of the bits in the different states is below the threshold.

15. The system of claim 12, further comprising instructions for performing an error correction code operation on the data associated with the codeword based at least in part on the data associated with the codeword being read from the memory address of the memory device.

16. The system of claim 12, further comprising instructions for: receiving a read request associated with a codeword located at a particular memory address of the memory device; determining whether the particular memory address is associated with a particular entry in the look-aside buffer; and Based at least in part on determining that the particular memory address is associated with the particular entry in the bypass buffer, data associated with the particular entry is returned.

17. The system of claim 12, further comprising instructions for: receiving a read request, the read request being associated with a codeword located at a particular memory address of the memory device; reading data associated with the codeword from the particular memory address of the memory device; determining whether the particular memory address is associated with a particular entry in the look-aside buffer; as well as Based at least in part on determining that the particular memory address is associated with the particular entry in the bypass buffer, data associated with the particular entry is returned.

18. The system of claim 12, further comprising instructions for: receiving a write request, the write request being associated with a particular memory address of the memory device; determining whether the particular memory address is associated with a particular entry in the look-aside buffer; as well as Based at least in part on determining that the particular memory address is associated with the particular entry in the bypass buffer, data associated with the write request is written to the particular entry in the bypass buffer.

19. A data storage device comprising: means for reading data associated with a codeword located at a memory address of a memory device of said data storage device; means for storing said data associated with said codeword in a temporary storage location; means for writing test data into said memory address; means for determining a number of worn bits associated with the codeword, the number of worn bits based at least in part on a number of bits having a first state associated with the test data and a number of bits having a second state different from the first state; means for storing the data associated with the codeword as an entry in a bypass buffer based at least in part on determining that the number of the bits in the second state is above a threshold; and Means for associating the entry in the bypass buffer with the memory address of the memory device.

20. The data storage device of claim 19, further comprising means for updating a valid entry indicator associated with the entry in the bypass buffer.