Storage controller and method of restoring error of the same

The storage controller efficiently recovers from errors by using an upper controller to restore the lower controller's state, ensuring rapid recovery and maintaining system functionality.

KR103044656B1Active Publication Date: 2026-11-06SAMSUNG ELECTRONICS CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
KR1020200162164
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-11-27
Publication Date
2026-11-06
Estimated Expiration
2040-11-27

AI Technical Summary

Technical Problem

Existing storage controllers face challenges in efficiently recovering from errors and restarting operations in lower controllers, leading to potential data loss or critical failures.

Method used

A storage controller architecture with an upper controller and a lower controller, where the upper controller performs error recovery by backing up and restoring the lower controller's state information, allowing efficient error recovery and restart without affecting the upper controller.

Benefits of technology

The solution enables rapid error recovery and restart of the storage controller within 200 ms, maintaining system functionality and data integrity by resetting only the lower controller.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112020128168136-PAT00001_ABST
    Figure 112020128168136-PAT00001_ABST
Patent Text Reader

Abstract

A storage controller includes a host interface for communicating with a host device, a memory interface for communicating with a non-volatile memory device, an upper controller, and a lower controller. The upper controller issues operations to be performed on the non-volatile memory device based on requests received through the host interface. The lower controller includes an operation memory that stores operation codes and operation data, and controls the memory interface to perform issued operations received from the upper controller based on the operation codes and operation data. In the event that an error occurs in the lower controller, the upper controller performs an error recovery operation to restore the lower controller to a state prior to the error occurring in the lower controller based on the lower controller's status information. By efficiently recovering the error through an error recovery operation based on the lower controller's status information when an error occurs in the lower controller, and by resetting only the lower controller, the storage controller can be efficiently restarted.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a semiconductor integrated circuit, and more specifically, to a storage controller and an error recovery method for a storage controller. Background Technology

[0002] Semiconductor memory devices are generally classified into volatile memory devices such as DRAM (dynamic random access memory) and SRAM (static random access memory), and non-volatile memory devices such as flash memory, FRAM (ferroelectric random access memory), PRAM (phase-change random access memory), and MRAM (magnetic random access memory). Volatile memory devices lose stored data when the power is cut off, whereas non-volatile memory devices retain stored data even when the power is cut off. In particular, flash memory, a type of non-volatile memory, has advantages such as fast programming speed, low power consumption, and large data storage capacity. Accordingly, flash memory is widely used as a storage medium in various fields requiring low-power and large-capacity storage devices, such as MP3 players, digital cameras, Solid State Drives (SSDs), embedded multimedia cards (eMMCs), and computer systems.

[0003] The storage controller controlling the above-mentioned non-volatile memory device includes various components. If an error occurs in some of these components, data stored in the non-volatile memory device may be lost or a critical failure of the storage device may occur. The problem to be solved

[0004] One objective of the present invention to solve the above-mentioned problems is to provide a storage controller capable of efficiently recovering errors in a lower controller and restarting it in a storage controller comprising an upper controller and a lower controller.

[0005] In addition, one objective of the present invention is to provide an error recovery method for a storage controller that can efficiently recover and restart an error in a lower controller in a storage controller comprising an upper controller and a lower controller. means of solving the problem

[0006] To achieve the above objective, a storage controller according to embodiments of the present invention includes a host interface for communicating with a host device, a memory interface for communicating with a non-volatile memory device, an upper controller, and a lower controller.

[0007] The above-mentioned upper controller issues operations to be performed in the non-volatile memory device based on requests received through the host interface.

[0008] The lower controller includes an operation memory that stores operation codes and operation data, and controls the memory interface to perform issuance operations received from the upper controller based on the operation codes and operation data.

[0009] The upper controller performs an error recovery operation to restore the lower controller to its state prior to the error occurring, based on the status information of the lower controller, when an error occurs in the lower controller.

[0010] To achieve the above objective, an error recovery method for a storage controller according to embodiments of the present invention comprises: a step of generating an interrupt when an error occurs in a lower controller that controls a memory interface for communicating with a non-volatile memory device to perform publishing operations received from an upper controller based on the operation code and the operation data, the lower controller including an operation memory that stores operation code and operation data; a step of stopping the operation of the lower controller and backing up the state information of the lower controller to the upper controller in response to the interrupt by the upper controller; and a step of performing an error recovery operation by the upper controller to restore the lower controller to a state prior to the error occurring in the lower controller based on the backed-up state information of the lower controller.

[0011] To achieve the above objective, an error recovery method for a storage controller according to embodiments of the present invention comprises: generating an interrupt when an error occurs in a lower controller that controls a memory interface that communicates with a non-volatile memory device to perform issuance operations received from an upper controller based on the operation code and the operation data, the lower controller including an operation memory that stores operation code and operation data; stopping the operation of the lower controller and the operation of the memory interface in response to the interrupt by the upper controller and backing up the state information of the lower controller and the state information of the memory interface to the upper controller; and performing an error recovery operation by the upper controller to restore the lower controller and the memory interface to a state prior to the error of the lower controller based on the backed-up state information of the lower controller and the backed-up state information of the memory interface. Effects of the invention

[0012] The storage controller and the error recovery method of the storage controller according to embodiments of the present invention can efficiently recover an error through an error recovery operation based on the status information of the lower controller when an error occurs in the lower controller, and efficiently restart the storage controller by resetting only the lower controller. Brief explanation of the drawing

[0013] FIG. 1 is a flowchart illustrating an error recovery method of a storage controller according to embodiments of the present invention. FIG. 2 is a block diagram showing a storage system according to embodiments of the present invention. FIG. 3 is a block diagram showing a storage controller according to embodiments of the present invention. FIG. 4 is a diagram showing an example of the hierarchical structure of the firmware of a storage controller according to embodiments of the present invention. FIG. 5 is a diagram showing an example of the firmware configuration of an upper controller and a lower controller included in a storage controller according to embodiments of the present invention. FIGS. 6 and 7 are drawings illustrating an example of an error recovery operation of a higher-level controller included in a storage controller according to embodiments of the present invention. FIGS. 8 and 9 are drawings illustrating other embodiments of an error recovery operation of a higher-level controller included in a storage controller according to embodiments of the present invention. FIG. 10 is a flowchart illustrating an error recovery method of a storage controller according to embodiments of the present invention. FIGS. 11 and 12 are drawings illustrating an example of an error recovery operation of a higher-level controller included in a storage controller according to embodiments of the present invention. FIG. 13 is a sequence diagram for explaining the operation of a storage controller according to embodiments of the present invention. FIG. 14 is a block diagram showing one embodiment of a non-volatile memory device included in a storage device according to embodiments of the present invention. FIG. 15 is a block diagram showing a memory cell array included in the non-volatile memory device of FIG. 14. FIG. 16 is a circuit diagram showing an equivalent circuit of a memory block included in the memory cell array of FIG. 15. FIG. 17 is a block diagram showing an electronic device according to embodiments of the present invention. Specific details for implementing the invention

[0014] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. Identical components in the drawings are given the same reference numerals, and redundant descriptions of identical components are omitted.

[0015] FIG. 1 is a flowchart illustrating an error recovery method of a storage controller according to embodiments of the present invention.

[0016] Referring to FIG. 1, an interrupt is generated when an error occurs in a lower controller that controls a memory interface that includes an operation memory storing operation codes and operation data and communicates with a non-volatile memory device to perform issuance operations received from an upper controller based on the operation codes and operation data (S100).

[0017] The above upper controller stops the operation of the lower controller in response to the interrupt and backs up the status information of the lower controller to the upper controller (S200).

[0018] Based on the state information of the lower controller backed up by the upper controller, an error recovery operation is performed to restore the lower controller to the state prior to the error occurring in the lower controller (S300).

[0019] Thus, the storage controller and the error recovery method of the storage controller according to the embodiments of the present invention can efficiently recover the error through an error recovery operation based on the status information of the lower controller when an error occurs in the lower controller, and efficiently restart the storage controller by resetting only the lower controller.

[0020] Hereinafter, embodiments relating to a storage controller according to embodiments of the present invention and an error recovery method of a storage controller according to embodiments of the present invention will be described in more detail with reference to FIGS. 2 and 3.

[0021] FIG. 2 is a block diagram showing a storage system according to embodiments of the present invention.

[0022] Referring to FIG. 2, the storage system (1000) may include a host device (2000) and a storage device (3000). The storage device (3000) may include a storage controller (100) and a non-volatile memory device (400). For example, the storage device (3000) may be a solid state drive (SSD), an embedded multimedia card (eMMC), a universal flash storage (UFS) device, etc.

[0023] The host device (2000) may be a data processing device capable of processing data, such as a CPU (central processing unit), a processor, a microprocessor, or an application processor. The storage device (3000) may be embedded in an electronic device together with the host device (2000), or may be electrically connected to an electronic device including the host device (2000) in a detachable manner.

[0024] The host device (2000) can transmit a data operation request, i.e., a request (REQ) and an address (ADDR), to the storage controller (100), and can exchange data (DTA) with the storage controller (100). The storage controller (100) can transmit a response (RSND) to the data operation request (REQ) to the host device (2000). The data operation request (REQ) may include a read request, a program request, and an erase request, etc. That is, the data operation may include read, program, and erase operations, etc.

[0025] The storage controller (100) can control the non-volatile memory device (400) in response to a request (REQ) from the host device (2000). The storage controller (100) can perform read operations, program operations, and erase operations on the non-volatile memory device (400) by providing an address (ADDR), a command (CMD), and a control signal (CTRL), etc., to the non-volatile memory device (400). The program operation may be called a write operation.

[0026] For example, the storage controller (100) may control the non-volatile memory device (400) to read out data (DTA) stored in the non-volatile memory device (400) in response to a read request received from the host device (2000), or to write data (DTA) to the non-volatile memory device (400) in response to a program request received from the host device (2000). Additionally, the non-volatile memory device (400) may control the non-volatile memory device (400) to erase data (DTA) stored in the non-volatile memory device (400) in response to an erase request received from the host device (2000). The non-volatile memory device (400) may transmit a response (RSND) to a command (CMD) to the storage controller (100).

[0027] The storage controller (100) may include an upper controller (HCON) (200) and a lower controller (LCON) (300).

[0028] The upper controller (200) can issue operations to be performed in the non-volatile memory device (400) based on a request (REQ) received from the host device (2000).

[0029] The lower controller (300) includes an operation memory that stores operation codes and operation data, and can control a non-volatile memory device (400) to perform issuance operations received from the upper controller based on the operation codes and operation data.

[0030] As described below with reference to FIG. 3, the storage controller (100) may include a host interface for communicating with a host device (2000) and a memory interface for communicating with a non-volatile memory device (400).

[0031] In one embodiment, as described below with reference to FIGS. 6 to 9, the upper controller (200) of the storage controller (100) can perform an error recovery operation to restore the lower controller (300) to a state prior to the error occurring in the lower controller (300) based on the state information of the lower controller (300) when an error occurs in the lower controller (300).

[0032] In another embodiment, as described below with reference to FIGS. 10 to 12, the upper controller (200) of the storage controller (100) can perform an error recovery operation to restore the lower controller (300) and the memory interface to the state prior to the error in the lower controller (300) occurring, based on the state information of the lower controller (300) as well as the state information of the memory interface, when an error in the lower controller (300) occurs.

[0033] The non-volatile memory device (400) may be implemented with one or more non-volatile memories (NVM), such as flash memory, MRAM (Magnetic RAM), FeRAM (Ferroelectric RAM), PRAM (Phase change RAM), and ReRAM (Resistive RAM). The non-volatile memory device (400) may be connected to a storage controller (100) through multiple channels (CH0~CHm). For convenience of explanation, the non-volatile memory device (400) may be described below as an example of a NAND flash memory device.

[0034] FIG. 3 is a block diagram showing a storage controller according to embodiments of the present invention. Descriptions that overlap with FIG. 2 may be omitted below.

[0035] Referring to FIGS. 2 and 3, the storage controller (100) may include a control unit (110), a host interface (HIF) (120), a ROM (Read Only Memory) (130), an internal memory (BUFF) (140), a memory interface (MIF) (150), and an internal bus (160) that electrically connects them.

[0036] The host interface (120) is connected to the host device (2000) via the bus (10) and can provide an interface function with the host device (2000). For example, the host device (2000) can exchange data with the host interface (120) based on at least one of various interface protocols such as USB (Universal Serial Bus) protocol, MMC (Multi Media Card) protocol, PCI (Peripheral Component Interconnection) protocol, PCI-E (PCI-Express) protocol, ATA (Advanced Technology Attachment) protocol, Serial-ATA protocol, Parallel-ATA protocol, SCSI (Small Computer Small Interface) protocol, SAS (serial attached SCSI) protocol, ESDI (Enhanced Small disk Interface) protocol, IDE (Integrated Drive Electronics) protocol, MIPI (Mobile Industry Processor Interface) protocol, and UFS (Universal Flash Storage) protocol.

[0037] The host interface (120) may include a host interface control logic (HICL) that controls the transmission and transfer of signals and data with the host device (2000). The host interface control logic (HICL) may include a register (HISFR) that stores setting values ​​for controlling the operation of the host interface (120) and status information indicating the current operation of the host interface (120).

[0038] The memory interface (150) is connected to the non-volatile memory device (400) via the bus (20) and can provide an interface function with the non-volatile memory device (400). For example, the bus (20) can be implemented with a plurality of channels (CH0~CHm) that can communicate with the non-volatile memory device (400) independently of each other, as described with reference to FIG. 2. According to an embodiment, the memory interface (150) can be implemented to comply with standard protocols such as Toggle or ONFI.

[0039] The memory interface (150) may include memory interface control logic (MICL) that controls the transmission and transfer of signals and data with a non-volatile memory device (400). The memory interface control logic (MICL) may include a register (MISFR) that stores setting values ​​for controlling the operation of the memory interface (150) and status information indicating the current operation of the memory interface (150). The memory interface control logic (MICL) may include FMC (flash memory control) logic, NSP (NAND signal processing) logic, RAID (Redundant Array of Independent Disk) logic, etc.

[0040] The ROM (130) can store code and data required for the operation of the storage controller (100). The internal memory (140) can buffer and store write data transmitted from the host device (2000) and read data transmitted from the non-volatile memory device (400). Additionally, during the initialization or booting process of the storage device (3000), code and data required for the operation of the storage controller (100) can be loaded into the internal memory (140), and metadata stored in the non-volatile memory device (400) can be loaded into the internal memory (140).

[0041] In FIG. 3, the ROM (130) and internal memory (140) are depicted as being contained within the storage controller (100), but are not necessarily limited thereto. Depending on the embodiment, the ROM (130) and internal memory (140) may exist separately outside the storage controller (100). In one embodiment, the internal memory (140) may be implemented as dynamic random access memory (DRAM).

[0042] The control unit (110) can control the overall operation of the storage controller (100). The control unit may include at least one of a central processing unit (CPU) and a microprocessing unit (MCU). The control unit (110) can run firmware for controlling the storage controller (100).

[0043] The control unit (110) may include a plurality of processor cores or a plurality of controllers. For example, the control unit (110) may include a host core (HCORE) (111), a Flash Translation Layer (FTL) core (FCORE) (112), and a NAND Manager (NM) core (NMCORE) (113). The host core (111), the FTL core (112), and the NM core (113) may each be implemented as separate processor cores.

[0044] The host core (111) can receive a set of requests input from the host interface (120). The host core (111) can parse the set of requests and distribute them to cores corresponding to each request (REQ).

[0045] The host core (111) may include an operation memory (HOM), parity logic (PTRC), and a register (HSFR). The parity logic (PTRC) may generate parity bits corresponding to operation codes and operation data for driving the host core (111) and store them in the operation memory (HOM) together with the operation codes and operation data. The register (HSFR) may store setting values ​​for controlling the operation of the host core (111) and state information indicating the current operation of the host core (111).

[0046] The FTL core (112) can control the NM core (113) to perform operations such as read operations, program operations, erase operations, etc., on the non-volatile memory device (400) based on requests (REQs) received from the host core (111).

[0047] The FTL core (112) may include an operation memory (FOM), parity logic (PTRC), and a register (FSFR). The parity logic (PTRC) may generate parity bits corresponding to the operation code and operation data for driving the FTL core (112) and store them in the operation memory (FOM) together with the operation code and the operation data. The register (FSFR) may store setting values ​​for controlling the operation of the FTL core (112) and state information indicating the current operation of the FTL core (112).

[0048] The NM core (113) can control the memory interface (150) so that operations on the non-volatile memory device (400) are performed according to the control of the FTL core (112).

[0049] The NM core (113) may include an operation memory (NMOM), parity logic (PTRC), and a register (NMSFR). The parity logic (PTRC) may generate parity bits corresponding to operation codes and operation data for driving the NM core (113) and store them in the operation memory (NMOM) together with the operation codes and operation data. The register (NMSFR) may store setting values ​​for controlling the operation of the NM core (113) and state information indicating the current operation of the NM core (113).

[0050] During the initialization process of the storage controller (100), operation codes and operation data required for the operation of the host core (111), FTL core (112), and NM core (113) may be loaded from the ROM (130) and / or non-volatile memory device (400) and stored in the internal memory (140). The host core (111), FTL core (112), and NM core (113) may load all or part of the operation codes and operation data required for the operation of each core from the operation codes and operation data stored in the internal memory (140) and store them in the operation memories (HOM, FOM, NMOM), respectively. According to embodiments, operation codes and operation data may be loaded from the ROM (130) and / or non-volatile memory device (400) and stored directly in the operation memories (HOM, FOM, NMOM). For example, operational memories (HOM, FOM, NMOM) can be implemented as static random access memory (SRAM).

[0051] As described above, the FTL core (112) issues operations to be performed in the non-volatile memory device (400) based on a request received through the host interface (120), and the NM core (113) controls the memory interface (150) to perform the operations issued from the FTL core (112) based on operation codes and operation data stored in the operation memory (NMOM). Accordingly, the FTL core (112) corresponds to a processor core or controller superior to the NM core (113), and below, the FTL core (112) may be referred to as the superior controller (FCORE) and the NM core (113) may be referred to as the inferior controller (NMCORE).

[0052] In one embodiment, as described below with reference to FIGS. 6 to 9, the upper controller (FCORE) can perform an error recovery operation to restore the lower controller (NMCORE) to a state prior to the error occurring in the lower controller (NMCORE) based on the state information of the lower controller (NMCORE) when an error occurs in the lower controller (NMCORE).

[0053] In another embodiment, as described below with reference to FIGS. 10 to 12, the upper controller (FCORE) can perform an error recovery operation to restore the lower controller (NMCORE) and the memory interface (150) to the state prior to the error in the lower controller (NMCORE) occurring, based on the state information of the lower controller (NMCORE) as well as the state information of the memory interface (150) when an error in the lower controller (NMCORE) occurs.

[0054] The parity logic (PTRC) of the lower controller (NMCORE) can perform a parity check on the operation code or operation data read from the operation memory (NMOM) and generate an interrupt if a parity error occurs. The upper controller (FCORE) can respond to the interrupt to stop the operation of the lower controller (NMCORE) and back up the status information of the lower controller (NMCORE) and the status information of the memory interface (150). The status information of the lower controller (NMCORE) may include the values ​​stored in the operation memory (NMOM) and register (NMSFR) of the lower controller (NMCORE) at the time when the operation of the lower controller (NMCORE) is stopped. The status information of the memory interface (150) may include the values ​​stored in the register (MISFR) of the memory interface (150) at the time when the operation of the memory interface (150) is stopped.

[0055] In this way, by performing an error recovery operation based on the status information of the lower controller (NMCORE) and the memory interface (150), only the lower controller (NMCORE) can be reset, and the upper controller (FCORE) can operate normally without being reset.

[0056] Although not illustrated in FIG. 3, the storage controller (100) may further include various components such as an ECC (error check code) engine, a RAID (Redundant Array of Independent Disk) engine, an AES (Advanced Encryption Standard) engine, a clock supply unit, etc.

[0057] The above ECC engine can perform an error correction function for data written to the non-volatile memory device (400) or data read from the non-volatile memory device (400).

[0058] The above RAID engine can perform the function of dividing data corresponding to one stripe set and dividing and storing the divided stripes in multiple locations of a non-volatile memory device (400).

[0059] The above AES engine can perform at least one of an encryption operation and a decryption operation for data written to the non-volatile memory device (400) or data read from the non-volatile memory device (400) using a symmetric-key algorithm. The above AES engine may include an encryption module and a decryption module. According to an embodiment, the encryption module and the decryption module may be implemented as separate modules or as a single module.

[0060] The above clock supply unit can distribute an external clock signal to generate clock signals used for the operation of each part of the storage controller (100).

[0061] FIG. 4 is a diagram showing an example of the hierarchical structure of the firmware of a storage controller according to embodiments of the present invention.

[0062] Referring to FIGS. 3 and 4, the firmware of the storage controller (100) may be composed of a Host Interface Layer (HIL), a Flash Translation Layer (FTL), a Virtual Flash Layer (VFL), a Flash Interface Layer (FIL), a Virtual Interrupt Control Layer (VIC), and a NAND Manager Layer.

[0063] When a request transmitted from a host device (2000) is received by the host interface (120), the host interface layer (HIL), which is executed by the host core (HCORE), can decode the request and write it to a register (HISFR). The host interface layer (HIL) receives the information written to the register (HISFR) via polling or interrupt and can determine whether the received request is suitable for the protocol.

[0064] The flash translation layer (FTL) performed by the upper controller (FCORE) can map a logical block address (LBA; Logical Block Address) transmitted from the host device (2000) to a physical block address (PBA; Physical Block Address), which is a physical location of the non-volatile memory device (400).

[0065] The NM layer performed by the lower controller (NMCORE) can control the operation of the memory interface (150).

[0066] Meanwhile, the Virtual Flash Layer (VFL) and Flash Interface Layer (FIL) performed by the upper controller (FCORE) can control the operation of the lower controller (NMCORE).

[0067] Additionally, the virtual flash layer (VFL) can manage bad block information of the non-volatile memory device (400). The virtual interrupt control layer (VIC), executed by the upper controller (FCORE), can handle exceptional situations occurring in the lower controller (NMCORE).

[0068] Embodiments of the present invention relate to a virtual flash layer (VFL) to an NM layer (NML), and the firmware configuration of a storage controller (100) according to embodiments of the present invention will be described in more detail below with reference to FIG. 5.

[0069] FIG. 5 is a diagram showing an example of the firmware configuration of an upper controller and a lower controller included in a storage controller according to embodiments of the present invention.

[0070] Referring to FIGS. 3, 4, and 5, the NM layer (NML) of the lower controller (NMCORE) may include an operation code (OCD) and operation data (ODT) for controlling the memory interface (150), and a publication queue list (IQLST) containing operations (OP1~OP4) issued from the upper controller (FCORE). The lower controller (NMCORE) may store the operation code (OCD) and operation data (ODT) in the operation memory (NMOM) of the lower controller (NMCORE) and execute the stored operation code. Meanwhile, the publication queue list (IQLST) may be stored in the operation memory (NMON) of the lower controller (NMCORE), and may be stored in the operation memory (NMON) by modifying a portion of the initially loaded operation data (ODT).

[0071] As described above, the parity logic (PTRC) of the lower controller (NMCORE) can generate parity bits corresponding to the operation code (OCD) and operation data (ODT) and store them in the operation memory (NMOM) together with the operation code (OCD) and operation data (ODT). The parity logic (PTRC) can perform a parity check on the operation code (OCD) or operation data (ODT) based on the parity bits read along with the operation code (OCD) or operation data (ODT), and generate an interrupt if a parity error occurs.

[0072] The virtual flash layer (VFL) of the upper controller (FCORE) may include an input / output queue (IOQ). The input / output queue (IOQ) may include a wait queue (WTQ), a cancel queue (CNQ), and an error queue (ERQ). The upper controller (FCORE) may store the input / output queue (IOQ) in the upper controller (FCORE)'s operation memory (FOM) and / or internal memory (140).

[0073] The Wait Queue (WTQ) may contain waiting operations to be sent to the Lower Controller (NMCORE). The Cancel Queue (CNQ) may contain operations returned from among the operations sent to the Flash Interface Layer (FIL). The Error Queue (ERQ) may contain operations related to bad blocks of the non-volatile memory device (400). According to an embodiment, the Wait Queue (WTQ) and the Cancel Queue (CNQ) may be combined into one.

[0074] The virtual interrupt control layer (VIC) of the upper controller (FCORE) may include an interrupt handler (INTRH). In one embodiment, the upper controller (FCORE) may store operation code and operation data corresponding to the interrupt handler (INTRH) in the operation memory (FOM) of the upper controller (FCORE) and execute the stored operation code. In another embodiment, the interrupt handler (INTRH) may be implemented in the form of hardware that is physically distinct from the upper controller (FCORE) or included within the upper controller (FCORE).

[0075] The interrupt handler (INTRH) can stop the operation of the lower controller (NMCORE) and the memory interface (150) in response to the interrupt generated from the lower controller (NMCORE), and the recovery handler (RCVH) can back up the status information of the lower controller (NMCORE) and the memory interface (150).

[0076] As described above, the status information of the lower controller (NMCORE) may include values ​​stored in the operation memory (NMOM) and register (NMSFR) of the lower controller (NMCORE) at the time when the operation of the lower controller (NMCORE) is stopped. The status information of the memory interface (150) may include values ​​stored in the register (MISFR) of the memory interface (150) at the time when the operation of the memory interface (150) is stopped.

[0077] The flash interface layer (VFL) of the upper controller (FCORE) may include a recovery handler (RCVH) and a communication queue (CMQ). The upper controller (FCORE) may store operation codes and operation data corresponding to the recovery handler (RCVH) in the operation memory (FOM) of the upper controller (FCORE) and execute the stored operation codes. Meanwhile, the upper controller (FCORE) may store the communication queue (CMQ) in the operation memory (FOM) and / or internal memory (140) of the upper controller (FCORE).

[0078] The communication queue (CMQ) may contain publication information regarding publication operations that have been completed and transmitted to the sub-controller (NMCORE). As described above, the non-volatile memory device (400) and the memory interface (150) may be connected through a plurality of channels (CH0~CHm) that can communicate independently of each other. In this case, the communication queue (CMQ) may include a plurality of sub-queues (0CHQ~mCHQ) that each store publication information regarding publication operations corresponding to the plurality of channels (CH0~CHm).

[0079] The recovery handler (RCVH) can classify the issuance operations into at least one reissuance operation and cancellation operations excluding the at least one reissuance operation based on the issuance information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150).

[0080] The recovery handler (RCVH) can reconstruct the issue queue list (IQLST) of the sub-controller (NMCORE) based on the issue information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150). Subsequently, the recovery handler (RCVH) can determine at least one reissue operation among the operations of the reconstructed issue queue list (IQLST) based on the issue information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150), and determine the remaining operations excluding the at least one reissue operation as cancel operations. The recovery handler (RCVH) can store the reissue operations classified in this manner in the reissue queue list (RIQLST) and store the cancel operations in the cancel queue list (CNQLST). The classification of reissue operations and cancel operations will be explained in more detail with reference to FIGS. 6 through 12.

[0081] The recovery handler (RCVH) initializes the sub-controller (NMCORE) and performs a reset operation that saves the operation code (OCD) and operation data (ODT) to the sub-controller (NMCORE)'s operation memory (NMON). After the reset operation of the sub-controller (NMCORE) is completed, it can transmit the re-issuance operations stored in the re-issuance queue list (RIQLST) to the sub-controller (NMCORE). Meanwhile, the recovery handler (RCVH) can store the cancellation operations stored in the cancellation queue list (CNQLST) in the input / output queue (IOQ).

[0082] Subsequently, the upper controller (FCORE) or lower controller (NMCORE) may initialize the memory interface (150) and store setting values ​​for controlling the operation of the memory interface (150) and state information corresponding to the previous state in the register (MISFR) of the memory interface (150). Here, the previous state refers to the state prior to the occurrence of an error in the lower controller (NMCORE), representing a state corresponding to the first re-issuance operation to be newly performed.

[0083] In this way, re-issuance operations for the non-volatile memory device (400) can be performed after the reset operation for the lower controller (NMCORE) and the memory interface (150) is completed.

[0084] FIGS. 6 and 7 are drawings illustrating an example of an error recovery operation of a higher-level controller included in a storage controller according to embodiments of the present invention.

[0085] FIGS. 6 and 7 illustrate an example in which the issue queue list (IQLST) of the sub-controller (NMCORE) includes the first to sixth issue operations (OP1~OP6) at the time an error occurs. In FIG. 6, the first state (AF) represents a state after the corresponding issue operation has been initiated for the non-volatile memory device (400), the second state (BF) represents a state before the corresponding issue operation has been initiated for the non-volatile memory device (400), and the third state (DN) represents a state after the corresponding issue operation has been completed. In FIG. 7, the first time point (Te) represents the time when an error occurs in the sub-controller (NMCORE), and the second time point (Tr) represents the time when the reset operation of the sub-controller (NMCORE) by the recovery handler (RCVH) is completed.

[0086] The upper controller (FCORE) can perform scheduling for waiting operations stored in the input / output queue (IOQ) and transmit publish operations to the lower controller (NMCORE) in an appropriate order. For example, the first publish operation (OP1) to the sixth publish operation (OP6) can be transmitted sequentially from the upper controller (FCORE) to the lower controller (NMCORE).

[0087] For example, it is assumed that an error occurs in the sub-controller (NMCORE) while the third issuance operation (OP3) is being performed, and that the second issuance operation (OP2) is a completed operation at the time the error occurs.

[0088] Referring to FIGS. 3 through 7, the recovery handler (RCVH) can reconstruct the issue queue list (IQLST) of the sub-controller (NMCORE) based on the issue information stored in the communication queue (CMQ) at the first time point (Te) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150). The issue queue list (IQLST) may sequentially include the first to sixth issue operations (OP1~OP6).

[0089] Subsequently, the recovery handler (RCVH) may determine the first, third, and fourth issuance operations (OP1, OP3, OP4) corresponding to the first state (AF) as re-issuance operations among the first to sixth issuance operations (OP1~OP6) of the issuance queue list (IQLST) at the first time point (Te) that has been reconstructed based on the issuance information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150). Meanwhile, the recovery handler (RCVH) may determine the fifth and sixth issuance operations (OP5, OP6) corresponding to the second state (BF) as cancel operations.

[0090] After the reset operation of the lower controller (NMCORE) is completed, the recovery handler (RCVH) transmits the first, third, and fourth issuance operations (OP1, OP3, OP4) corresponding to the re-issuance operations to the lower controller (NMCORE), and the lower controller (NMCORE) can store the issuance queue list (IQLST) of the second time point (Tr). At this time, the upper controller (FCORE) first transmits the third issuance operation (OP3) in which an error occurred to the lower controller (NMCORE), and the lower controller (NMCORE) can restart communication with the non-volatile memory device (400) starting from the third issuance operation (OP3). Meanwhile, the recovery handler (RCVH) can store the fifth and sixth issuance operations (OP5, OP6) corresponding to the cancellation operations stored in the cancellation queue list (CNQLST) back into the input / output queue (IOQ).

[0091] FIGS. 8 and 9 are drawings illustrating other embodiments of an error recovery operation of a higher-level controller included in a storage controller according to embodiments of the present invention. Since the embodiments of FIGS. 8 and 9 are similar to the embodiments of FIGS. 6 and 7, redundant descriptions are omitted.

[0092] Referring to FIGS. 3, 4, 5, 8 and 9, the recovery handler (RCVH) can reconstruct the issue queue list (IQLST) of the sub-controller (NMCORE) based on the issue information stored in the communication queue (CMQ) at the first time point (Te) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150). The issue queue list (IQLST) may sequentially include the first to sixth issue operations (OP1 to OP6).

[0093] Subsequently, the recovery handler (RCVH) may determine the third issuance operation (OP3) in which an error occurred as a re-issuance operation among the first to sixth issuance operations (OP1~OP6) of the issuance queue list (IQLST) at the first time point (Te) that has been reconstructed based on the issuance information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150). Meanwhile, the recovery handler (RCVH) may determine the first, fourth, fifth, and sixth issuance operations (OP1, OP4, OP5, OP6) as cancel operations, excluding the third issuance operation (OP3) determined as a re-issuance operation and the second issuance operation (OP2) corresponding to the third state (DN).

[0094] After the reset operation of the sub-controller (NMCORE) is completed, the recovery handler (RCVH) transmits the third issuance operation (OP3), which corresponds to the reissuance operations, to the sub-controller (NMCORE), and the sub-controller (NMCORE) can store the issuance queue list (IQLST) of the second time point (Tr). Meanwhile, the recovery handler (RCVH) can store the first, fourth, fifth, and sixth issuance operations (OP1, OP4, OP5, OP6), which correspond to the cancellation operations stored in the cancellation queue list (CNQLST), back into the input / output queue (IOQ).

[0095] In this way, when an error occurs in the lower controller (NMCORE), the storage controller (100) can be made to operate normally again from the operation where the error occurred through the error recovery operation of the upper controller (FCORE).

[0096] The error recovery method of the storage controller (100) according to the embodiments of the present invention can be performed within a maximum of 200 ms and can improve the user experience of the storage system (1000) including the storage device (3000) through rapid recovery. In addition, the error recovery method of the storage controller (100) according to the embodiments of the present invention restores the sub-controller (NMCORE) and the memory interface (150) to the state prior to the error, thereby allowing the continued use of data that could not be used when an SRAM parity error occurred in the existing sub-controller (NMCORE).

[0097] FIG. 10 is a flowchart illustrating an error recovery method of a storage controller according to embodiments of the present invention.

[0098] Referring to FIG. 10, an interrupt is generated when an error occurs in a lower controller that controls a memory interface that includes an operation memory storing operation codes and operation data and communicates with a non-volatile memory device to perform issuance operations received from an upper controller based on the operation codes and operation data (S100).

[0099] The upper controller stops the operation of the lower controller and the operation of the memory interface in response to the interrupt, and backs up the status information of the lower controller and the status information of the memory interface to the upper controller (S210).

[0100] Based on the state information of the backed-up lower controller and the state information of the backed-up memory interface by the upper controller, an error recovery operation is performed to restore the lower controller and the memory interface to the state prior to the error of the lower controller occurring (S310).

[0101] In this way, the storage controller and the error recovery method of the storage controller according to the embodiments of the present invention can further subdivide the stage in which an error in a re-issuance operation occurs based on the state information of the lower controller (300) as well as the state information of the memory interface, as described below with reference to FIGS. 11 and 12, when an error in the lower controller (300) occurs.

[0102] Hereinafter, with reference to FIGS. 11 and 12, embodiments are described in which at least one operation is a multi-stage operation comprising a plurality of sub-operations that are sequentially performed on a non-volatile memory device (400).

[0103] FIGS. 11 and 12 are drawings illustrating an example of an error recovery operation of a higher-level controller included in a storage controller according to embodiments of the present invention.

[0104] FIGS. 11 and 12 illustrate an example in which the issue queue list (IQLST) of the sub-controller (NMCORE) includes first to third operations (OP1 to OP3) at the time an error occurs. For example, the first operation (OP1) may include first and second sub-operations (OP11, OP12) performed sequentially on the non-volatile memory device (400), the second operation (OP2) may include third, fourth, and fifth sub-operations (OP21, OP22, OP23) performed sequentially on the non-volatile memory device (400), and the third operation (OP3) may include sixth and seventh sub-operations (OP31, OP32) performed sequentially on the non-volatile memory device (400).

[0105] In FIG. 11, the first state (AF) represents a state after a corresponding sub-operation has been initiated for the non-volatile memory device (400), the second state (BF) represents a state before a corresponding sub-operation has been initiated for the non-volatile memory device (400), and the third state (DN) represents a state after a corresponding sub-operation has been completed. In FIG. 12, the first time point (Te) represents a time when an error occurred in the sub-controller (NMCORE), and the second time point (Tr) represents a time when the reset operation of the sub-controller (NMCORE) by the recovery handler (RCVH) is completed.

[0106] The upper controller (FCORE) can perform scheduling on operations stored in the input / output queue (IOQ) and transmit sub-operations of the issued operations to the lower controller (NMCORE) in an appropriate order. For example, they can be transmitted sequentially from the upper controller (FCORE) to the lower controller (NMCORE) in the order of the first sub-operation (OP11), the third sub-operation (OP21), the fourth sub-operation (OP22), the second sub-operation (OP12), the sixth sub-operation (OP31), the fifth sub-operation (OP23), and the seventh sub-operation (OP32).

[0107] In this way, a sub-operation belonging to another issuance operation may be initiated before all sub-operations belonging to one issuance operation are completed. For example, as illustrated in FIG. 12, depending on the scheduling of the upper controller (FCORE), a third sub-operation (OP21) corresponding to the first of the second operation (OP2) may be initiated before a second sub-operation (OP12) corresponding to the last of the first issuance operation (OP1).

[0108] Interleaving of sub-operations belonging to these different publishing operations may be possible when the non-volatile memory device (400) is divided into multiple channels and the channels corresponding to the publishing operations are different, or when the non-volatile memory device (400) supports per-plane operation and the memory planes corresponding to the publishing operations are different.

[0109] For example, it is assumed that an error occurs in the sub-controller (NMCORE) while performing the fourth sub-operation (OP22) included in the second issuance operation (OP2), and at the time the error occurs, the first sub-operation (OP11) included in the first issuance operation (OP1) and the third sub-operation (OP21) included in the second issuance operation (OP2) are completed sub-operations. The sub-operation where the error occurred may be referred to as the target sub-operation.

[0110] Referring to FIGS. 3, 4, 5, 11 and 12, the recovery handler (RCVH) can reconstruct the issue queue list (IQLST) of the sub-controller (NMCORE) based on the issue information stored in the communication queue (CMQ) at the first time point (Te) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150). The issue queue list (IQLST) may sequentially include the first sub-operation (OP11), the third sub-operation (OP21), the fourth sub-operation (OP22), the second sub-operation (OP12), the sixth sub-operation (OP31), the fifth sub-operation (OP23), and the seventh sub-operation (OP32).

[0111] Subsequently, the recovery handler (RCVH) can determine the fourth sub-operation (OP22) corresponding to the first state (AF) as the re-issuance sub-operation among the first to seventh sub-operations (OP11~OP32) of the reconstructed issuance queue list (IQLST) of the first time point (Te) based on the issuance information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150).

[0112] Additionally, the recovery handler (RCVH) may determine the second sub-operation (OP12) and the fifth sub-operation (OP23), which correspond to the second state (BF), as re-issuance sub-operations, which are included in the first and second issuance operations that include sub-operations (OP11, OP22) that have already been initiated or completed for the non-volatile memory device (400).

[0113] Meanwhile, the recovery handler (RCVH) can determine the 6th and 7th sub-operations (OP31, OP32) of the 3rd issuance operation (OP3) that has not yet been initiated for the non-volatile memory device (400) as cancel sub-operations.

[0114] After the reset operation of the sub-controller (NMCORE) is completed, the recovery handler (RCVH) transmits the second, fourth, and fifth sub-operations (OP12, OP22, OP23) corresponding to the reissue sub-operations to the sub-controller (NMCORE), and the sub-controller (NMCORE) can store the issue queue list (IQLST) of the second time point (Tr).

[0115] At this time, the upper controller (FCORE) first transmits the fourth sub-operation (OP22) in which an error occurred to the lower controller (NMCORE), and the lower controller (NMCORE) can resume communication with the non-volatile memory device (400) starting from the fourth sub-operation (OP22). Meanwhile, the recovery handler (RCVH) can store the sixth and seventh sub-operations (OP31, OP32), i.e., the third issuance operation (OP3), corresponding to the canceled sub-operations stored in the cancel queue list (CNQLST), back into the input / output queue (IOQ).

[0116] In this way, if the publishing operation where an error occurred is a multi-stage operation, the state of the sub-controller (NMCORE) can be restored so that the multi-stage operation is executed starting from the target sub-operation where the error occurred. By subdividing the sub-operation or stage where the error occurred in the re-publishing operation and restoring the sub-controller (NMCORE) and the memory interface (150) to the state prior to the error, so that the operation is executed starting from the target sub-operation where the error occurred, the latency between the host device (2000) and the storage device (3000) caused by the error can be further reduced, thereby further improving the user experience.

[0117] FIG. 13 is a sequence diagram for explaining the operation of a storage controller according to embodiments of the present invention.

[0118] Referring to FIG. 13, the parity logic (PTRC) of the lower controller (NMCORE) performs a parity check on the read operation code or operation data (S10) and can generate an interrupt if a parity error occurs (S11).

[0119] In response to the above interrupt, the interrupt handler (INTRH) of the upper controller (FCORE) is initiated (S12), and the interrupt handler (INTRH) can stop the operation of the lower controller (NMCORE) and the memory interface (150) (S13).

[0120] Afterwards, the recovery handler (RCVH) of the upper controller (FCORE) is initiated (S14), and the recovery handler (RCVH) can back up the status information of the lower controller (NMCORE) and the memory interface (150) (S15).

[0121] As described above, the recovery handler (RCVH) can reconstruct the issuance operations of the sub-controller (NMCORE) based on the issuance information stored in the communication queue (CMQ) and the status information of the backed-up sub-controller (NMCORE) and memory interface (150), and classify the issuance operations into at least one re-issuance operation and cancellation operations excluding the at least one re-issuance operation.

[0122] The recovery handler (RCVH) initializes the sub-controller (NMCORE) and performs a reset operation that stores the operation code (OCD) and operation data (ODT) in the operation memory (NMON) of the sub-controller (NMCORE) (S16), and after the reset operation of the sub-controller (NMCORE) is completed, it can transmit at least one re-issuance operation stored in the re-issuance queue list (RIQLST) to the sub-controller (NMCORE) (S17).

[0123] Afterwards, the upper controller (FCORE) (or lower controller (NMCORE)) can perform a reset operation to initialize the memory interface (150) and store setting values ​​for controlling the operation of the memory interface (150) and state information corresponding to the previous state in the register (MISFR) of the memory interface (150) (S18).

[0124] After such error recovery operation is completed, the storage controller (100) can initiate a reissue operation on the non-volatile memory device (400).

[0125] The storage controller and the error recovery method of the storage controller according to embodiments of the present invention can efficiently recover an error through an error recovery operation based on the status information of the lower controller when an error occurs in the lower controller, and efficiently restart the storage controller by resetting only the lower controller.

[0126] Hereinafter, with reference to FIGS. 14 to 16, a non-volatile memory device included in a storage device according to embodiments of the present invention will be described. Although a NAND flash memory device is described as an example of a non-volatile memory device, the embodiments of the present invention are not limited thereto.

[0127] FIG. 14 is a block diagram showing one embodiment of a non-volatile memory device included in a storage device according to embodiments of the present invention.

[0128] Referring to FIG. 14, the non-volatile memory device (800) may include a memory cell array (500), an address decoder (530), a page buffer circuit (510), a data input / output circuit (520), a control circuit (550), and a voltage generator (560). In one embodiment, the non-volatile memory device (800) may have a Cell Over Periphery (COP) structure in which a memory cell array is formed over a peripheral area (PREG) where peripheral circuits are formed. In this case, the memory cell array (500) is formed in a cell area (CREG), and the address decoder (530), page buffer circuit (510), data input / output circuit (520), control circuit (550), and voltage generator (560) may be formed in the peripheral circuit area (PREG) of FIG. 1.

[0129] The memory cell array (500) can be connected to an address decoder (530) through a plurality of string select lines (SSL), a plurality of word lines (WL), and a plurality of ground select lines (GSL). Additionally, the memory cell array (500) can be connected to a page buffer circuit (510) through a plurality of bit lines (BL). The memory cell array (500) may include a plurality of memory cells connected to a plurality of word lines (WL) and a plurality of bit lines (BL).

[0130] In one embodiment, the memory cell array (500) may be a three-dimensional memory cell array formed in a three-dimensional structure (or vertical structure) on a substrate. In this case, the memory cell array (500) may include cell strings comprising a plurality of memory cells formed by stacking each other, for example, vertical memory NAND strings.

[0131] The control circuit (550) receives a command signal (CMD) and an address signal (ADDR) from an external memory controller and can control the erase loop, program loop, and read operation of the non-volatile memory device (10) based on the command signal (CMD) and the address signal (ADDR). Here, the program loop may include a program operation and a program verification operation, and the erase loop may include an erase operation and an erase verification operation. Here, the read operation may include a normal read operation and a data recovery read operation.

[0132] For example, the control circuit (550) can generate control signals (CTL) for controlling the voltage generator (560) and control signals (PBC) for controlling the page buffer circuit (510) based on the command signal (CMD), and can generate a row address (R_ADDR) and a column address (C_ADDR) based on the address signal (ADDR). The control circuit (550) can provide the row address (R_ADDR) to the address decoder (530) and provide the column address (C_ADDR) to the data input / output circuit (520).

[0133] The address decoder (530) can be connected to the memory cell array (500) through a plurality of string select lines (SSL), a plurality of word lines (WL) and a plurality of ground select lines (GSL).

[0134] When a program operation or a read operation occurs, the address decoder (530) can determine one of the multiple word lines (WL) as a selected word line based on the row address (R_ADDR) provided by the control circuit (550), and determine the remaining word lines as non-selected word lines.

[0135] Additionally, during a program operation or read operation, the address decoder (530) can determine one of the multiple string selection lines (SSL) as a selected string selection line based on the row address (R_ADDR) provided by the control circuit (550), and determine the remaining string selection lines as non-selected string selection lines.

[0136] Additionally, during a program operation or read operation, the address decoder (530) can determine one of the multiple ground selection lines (GSL) as a selected ground selection line based on the row address (R_ADDR) provided by the control circuit (550), and determine the remaining ground selection lines as unselected ground selection lines.

[0137] The voltage generator (560) can generate word line voltages (VWL) required for the operation of the non-volatile memory device (30) based on control signals (CTL) provided from the control circuit (550). The word line voltages (VWL) generated from the voltage generator (560) can be applied as driving voltages to a plurality of word lines (WL) through the address decoder (530).

[0138] For example, during an erase operation, the voltage generator (560) may apply an erase voltage to the wells and / or common source lines of the memory block and, based on the erase address, apply an erase allowance voltage (e.g., ground voltage) to all word lines of the memory block or to word lines corresponding to some sub-blocks. During an erase verification operation, the voltage generator (560) may apply an erase verification voltage to all word lines of a single memory block or apply an erase verification voltage on a word-line basis.

[0139] For example, during program operation, the voltage generator (560) may apply a program voltage to the selected word lines and a program pass voltage to the unselected word lines. Additionally, during program verification operation, the voltage generator (560) may apply a program verification voltage to the selected word lines and a verification pass voltage to the unselected word lines.

[0140] Additionally, during a normal readout operation, the voltage generator (560) may apply a readout voltage to the selected word line and a readout pass voltage to the non-selected word lines. Additionally, during a data recovery readout operation, the voltage generator (560) may apply a readout voltage to the word line adjacent to the selected word line and a recovery readout voltage to the selected word line.

[0141] The page buffer circuit (510) may be connected to the memory cell array (500) through a plurality of bit lines (BL). The page buffer circuit (510) may include a plurality of page buffers. In one embodiment, one bit line may be connected to one page buffer. In another embodiment, two or more bit lines may be connected to one page buffer. The page buffer circuit (510) may temporarily store data to be programmed on a selected page during a program operation, and may temporarily store data read from a selected page during a read operation.

[0142] The data input / output circuit (520) can be connected to the page buffer circuit (510) via data lines (DL). During a program operation, the data input / output circuit (520) receives program data (DATA) from the memory controller (20) and can provide the program data (DATA) to the page buffer circuit (510) based on the column address (C_ADDR) provided by the control circuit (550). During a read operation, the data input / output circuit (520) can provide the read data (DATA) stored in the page buffer circuit (510) to the memory controller (20) based on the column address (C_ADDR) provided by the control circuit (550).

[0143] Additionally, the page buffer circuit (510) and the input / output circuit (520) can read data from a first storage area of ​​the memory cell array (500) and write the read data to a second storage area of ​​the memory cell array (500). That is, the page buffer circuit (510) and the input / output circuit (520) can perform a copy-back operation. The page buffer circuit (510) and the input / output circuit (520) can be controlled by the control circuit (550).

[0144] FIG. 15 is a block diagram showing a memory cell array included in the non-volatile memory device of FIG. 14, and FIG. 16 is a circuit diagram showing an equivalent circuit of a memory block included in the memory cell array of FIG. 15.

[0145] As illustrated in FIG. 15, the memory cell array (500) may include a plurality of memory blocks (BLK1 to BLKz). The memory blocks (BLK1 to BLKz) are selected by an address decoder (530). For example, the address decoder (530) may select a memory block corresponding to a block address among the memory blocks (BLK1 to BLKz).

[0146] The memory block (BLKi) illustrated in FIG. 16 represents a three-dimensional memory block formed in a three-dimensional structure on a substrate. For example, a plurality of memory NAND strings included in the memory block (BLKi) may be formed in a direction (D3) perpendicular to the substrate.

[0147] Referring to FIG. 16, a memory block (BLKi) may include a plurality of cell strings, i.e., a plurality of memory NAND strings (NS11 to NS33), connected between bit lines (BL1, BL2, BL3) and a common source line (CSL). Each of the plurality of memory NAND strings (NS11 to NS33) may include a string select transistor (SST), a plurality of memory cells (MC1, MC2, ..., MC8), and a ground select transistor (GST). FIG. 16 illustrates that each of the plurality of memory NAND strings (NS11 to NS33) includes eight memory cells (MC1, MC2, ..., MC8), but embodiments of the present invention are not limited thereto.

[0148] A string select transistor (SST) can be connected to corresponding string select lines (SSL1, SSL2, SSL3). Multiple memory cells (MC1, MC2, ..., MC8) can each be connected to corresponding gate lines (GTL1, GTL2, ..., GTL8). The gate lines (GTL1, GTL2, ..., GTL8) may correspond to word lines, and some of the gate lines (GTL1, GTL2, ..., GTL8) may correspond to dummy word lines. A ground select transistor (GST) can be connected to corresponding ground select lines (GSL1, GSL2, GSL3). A string select transistor (SST) is connected to corresponding bit lines (BL1, BL2, BL3), and a ground select transistor (GST) can be connected to a common source line (CSL).

[0149] Word lines of the same height (e.g., GTL1) are connected in common, and ground select lines (GSL1, GSL2, GSL3) and string select lines (SSL1, SSL2, SSL3) can be separated. FIG. 16 is illustrated as a memory block (BLK) connected to eight gate lines (GTL1, GTL2, ..., GTL8) and three bit lines (BL1, BL2, BL3), but embodiments of the present invention are not limited thereto.

[0150] FIG. 17 is a block diagram showing an electronic device according to embodiments of the present invention.

[0151] Referring to FIG. 17, the electronic device (4000) includes an application processor (4100), a communication module (4200), a display / touch module (4300), a storage device (4400), and a buffer RAM (4500).

[0152] The application processor (4100) controls the overall operation of the electronic device (4000). The application processor (4100) can execute applications that provide an internet browser, games, videos, etc. The communication module (4200) can be implemented to control wired communication and / or wireless communication with the outside. The display / touch module (4300) can be implemented to display data processed by the application processor (4100) or to receive data input from a touch panel. The storage device (4400) can be implemented to store user data.

[0153] The storage device (4400) may be an embedded multimedia card (eMMC), a solid state drive (SSD), or a universal flash storage (UFS) device. As described above, the storage device (4400) includes an upper controller (FCORE) and a lower controller (NMCORE), and in the event of an error in the lower controller (NMCORE), the lower controller (NMCORE) can be restored to its state prior to the error under the control of the upper controller (FCORE).

[0154] The buffer RAM (4500) may be implemented to temporarily store data required during the processing operation of the electronic device (4000). For example, the buffer RAM (4500) may be a dynamic random access memory such as DDR SDRAM, LPDDR SDRAM, GDDR SDRAM, RDRAM, etc.

[0155] As described above, the storage controller and the error recovery method of the storage controller according to the embodiments of the present invention can efficiently recover an error through an error recovery operation based on the status information of the lower controller when an error occurs in the lower controller, and efficiently restart the storage controller by resetting only the lower controller.

[0156] A person skilled in the art will understand that the error recovery method according to embodiments of the present invention may be implemented in the form of a system, a method, a product including computer-readable program code stored on a computer-readable medium, etc. The computer-readable program code may be provided to a processor of various computers or other data processing devices. The computer-readable medium may be a computer-readable signal medium or a computer-readable recording medium. The computer-readable recording medium may be any tangible medium capable of storing or containing a program within or connected to an instruction execution system, equipment, or device. Industrial applicability

[0157] Embodiments of the present invention can be usefully utilized in storage devices and systems including the same. In particular, embodiments of the present invention can be more usefully applied to electronic devices such as memory cards, solid state drives (SSDs), embedded multimedia cards (eMMCs), universal flash storage (UFS), computers, laptops, cellular phones, smartphones, MP3 players, Personal Digital Assistants (PDAs), Portable Multimedia Players (PMPs), digital TVs, digital cameras, portable game consoles, navigation devices, wearable devices, Internet of Things (IoT) devices, Internet of Everything (IoE) devices, e-books, virtual reality (VR) devices, augmented reality (AR) devices, server systems, autonomous driving devices, etc.

[0158] Although the present invention has been described above with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as set forth in the following claims.

Claims

Claim 1 A storage controller comprising: a host interface for communicating with a host device; a memory interface for communicating with a non-volatile memory device; an upper controller for issuing operations to be performed on the non-volatile memory device based on a request received through the host interface; and a lower controller for controlling the memory interface to perform issuing operations received from the upper controller based on the operation code and the operation data, wherein the upper controller performs an error recovery operation to restore the lower controller to a state prior to the error occurring in the lower controller based on the state information of the lower controller when an error occurs in the lower controller, and the lower controller performs a parity check on the operation code or the operation data read from the operation memory and generates an interrupt when a parity error occurs, and the upper controller stops the operation of the lower controller and backs up the state information of the lower controller in response to the interrupt. Claim 2 A storage controller according to claim 1, characterized in that the upper controller and the lower controller are each implemented as separate processor cores. Claim 3 A storage controller according to claim 1, characterized in that, during the error recovery operation, only the lower controller is reset and the upper controller operates normally without being reset. Claim 4 A storage controller according to claim 1, wherein the lower controller generates parity bits corresponding to the operation code and the operation data and stores them in the operation memory together with the operation code and the operation data, and performs a parity check on the operation code or the operation data based on the parity bits read out together with the operation code or the operation data, and generates an interrupt when a parity error occurs. Claim 5 delete Claim 6 A storage controller according to claim 1, wherein the upper controller stores a communication queue that stores publishing information regarding publishing operations that have been completed to the lower controller, and determines at least one re-publishing operation to be transmitted to the lower controller among the publishing operations based on the publishing information stored in the communication queue and the backed-up status information of the lower controller. Claim 7 A storage controller according to claim 6, wherein the upper controller performs a reset operation to initialize the lower controller and then save the operation code and the operation data back to the lower controller, transmits the re-issuance operations to the lower controller after the reset operation of the lower controller is completed, and the lower controller performs the at least one re-issuance operation received from the upper controller after the reset operation is completed. Claim 8 A storage controller according to claim 6, wherein the upper controller determines the remaining operations among the issuance operations, excluding at least one re-issuance operation, as cancel operations based on the issuance information stored in the communication queue and the backed-up status information of the lower controller. Claim 9 A storage controller according to claim 8, wherein the upper controller stores an input / output queue containing waiting operations to be transmitted to the lower controller, and stores the cancel operations in the input / output queue. Claim 10 A storage controller according to claim 6, characterized in that the upper controller stops the operation of the memory interface in response to the interrupt and backs up the state information of the memory interface. Claim 11 A storage controller according to claim 10, wherein the error-causing issuance operation is a multi-stage operation comprising a plurality of sub-operations sequentially performed on the non-volatile memory device, and the upper controller determines the target sub-operation where the error occurred among the plurality of sub-operations of the multi-stage operation based on the state information of the backed-up memory interface. Claim 12 A storage controller according to claim 11, wherein the upper controller performs a reset operation to store the operation code and the operation data back into the lower controller after initializing the lower controller and the memory interface, transmits uncompleted re-issuance sub-operations among the plurality of sub-operations of the multi-stage operation to the lower controller, and the lower controller performs the re-issuance sub-operations received from the upper controller after the reset operation is completed. Claim 13 A storage controller according to claim 11, wherein the upper controller restores the state of the lower controller so that the lower controller performs the multi-stage operation from the target sub-operation where an error occurred. Claim 14 A storage controller error recovery method comprising: a step of generating an interrupt when an error occurs in a lower controller that controls a memory interface for communicating with a non-volatile memory device to perform issuance operations received from an upper controller based on the operation code and the operation data, the lower controller including an operation memory that stores operation codes and operation data; a step of stopping the operation of the lower controller in response to the interrupt and backing up the state information of the lower controller to the upper controller by the upper controller; and a step of performing an error recovery operation by the upper controller to restore the lower controller to a state prior to the error of the lower controller based on the backed-up state information of the lower controller. Claim 15 An error recovery method for a storage controller in claim 14, characterized in that, during the error recovery operation, only the lower controller is reset and the upper controller operates normally without being reset. Claim 16 An error recovery method for a storage controller according to claim 14, wherein the step of generating the interrupt comprises: a step of performing a parity check on the operation code or operation data read from the operation memory; and a step of generating the interrupt when a parity error occurs as a result of performing the parity check. Claim 17 An error recovery method for a storage controller according to claim 14, wherein the step of performing the error recovery operation comprises: storing a communication queue that stores publication information regarding the publication operations that have been completed to be transmitted to the sub-controller; and determining cancellation operations to be stored in an input / output queue including re-publication operations to be transmitted to the sub-controller and waiting operations to be transmitted to the sub-controller among the publication operations, based on the publication information stored in the communication queue and the backed-up status information of the sub-controller. Claim 18 An error recovery method for a storage controller according to claim 17, further comprising the step of stopping the operation of the memory interface and backing up the state information of the memory interface by the upper controller in response to the interrupt. Claim 19 In claim 18, the re-issuance operations include at least one multi-stage operation comprising a plurality of sub-operations sequentially performed on the non-volatile memory device, and the step of performing an error recovery operation comprises: determining, by the upper controller, a target sub-operation where an error occurred among the plurality of sub-operations of the multi-stage operation based on state information of the backed-up memory interface; performing a reset operation by the upper controller to initialize the lower controller and the memory interface and then store the operation code and the operation data back into the lower controller; transmitting the re-issuance operations from the upper controller to the lower controller; and performing the re-issuance operations received from the upper controller by the lower controller after the reset operation is completed. Claim 20 A storage controller error recovery method comprising: a step of generating an interrupt when an error occurs in a lower controller that controls a memory interface, which includes an operation memory storing operation codes and operation data, and communicates with a non-volatile memory device to perform issuance operations received from an upper controller based on the operation codes and operation data; a step of stopping the operation of the lower controller and the operation of the memory interface in response to the interrupt by the upper controller and backing up the state information of the lower controller and the state information of the memory interface to the upper controller; and a step of performing an error recovery operation by the upper controller to restore the lower controller and the memory interface to a state prior to the error of the lower controller based on the backed-up state information of the lower controller and the backed-up state information of the memory interface.

Citation Information

Patent Citations

  • Nonvolatile memory system and the operarion method there-of

    KR1020110132202A

  • Error handling method, associated data storage device and controller thereof

    KR1020200038857A

  • Reset and error handling in application specific integrated circuits

    US20190079817A1