WO2022226997A1 - 存储单元的访问方法、修复方法、裸片和存储芯片 - Google Patents
存储单元的访问方法、修复方法、裸片和存储芯片 Download PDFInfo
- Publication number
- WO2022226997A1 WO2022226997A1 PCT/CN2021/091480 CN2021091480W WO2022226997A1 WO 2022226997 A1 WO2022226997 A1 WO 2022226997A1 CN 2021091480 W CN2021091480 W CN 2021091480W WO 2022226997 A1 WO2022226997 A1 WO 2022226997A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- unit
- redundant
- storage
- address
- failed
- Prior art date
Links
- 230000008439 repair process Effects 0.000 title claims abstract description 237
- 238000000034 method Methods 0.000 title claims abstract description 107
- 238000013500 data storage Methods 0.000 claims description 56
- 230000004044 response Effects 0.000 claims description 33
- 238000005538 encapsulation Methods 0.000 claims description 19
- 230000002950 deficient Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 description 33
- 238000013507 mapping Methods 0.000 description 21
- 238000010586 diagram Methods 0.000 description 20
- 238000012360 testing method Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 239000000047 product Substances 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 4
- 230000006386 memory function Effects 0.000 description 3
- 238000004806 packaging method and process Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 239000008187 granular material Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/76—Masking faults in memories by using spares or by reconfiguring using address translation or modifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
- G11C29/44—Indication or identification of errors, e.g. for repair
- G11C29/4401—Indication or identification of errors, e.g. for repair for self repair
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/72—Masking faults in memories by using spares or by reconfiguring with optimized replacement algorithms
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
- G11C29/785—Masking faults in memories by using spares or by reconfiguring using programmable devices with redundancy programming schemes
- G11C29/787—Masking faults in memories by using spares or by reconfiguring using programmable devices with redundancy programming schemes using a fuse hierarchy
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
- G11C2029/1202—Word line control
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
- G11C2029/4402—Internal storage of test result, quality data, chip identification, repair information
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
- G11C29/781—Masking faults in memories by using spares or by reconfiguring using programmable devices combined in a redundant decoder
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
- G11C29/785—Masking faults in memories by using spares or by reconfiguring using programmable devices with redundancy programming schemes
- G11C29/789—Masking faults in memories by using spares or by reconfiguring using programmable devices with redundancy programming schemes using non-volatile cells or latches
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
- G11C29/80—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved layout
- G11C29/808—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved layout using a flexible replacement scheme
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
- G11C29/80—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved layout
- G11C29/816—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved layout for an application-specific layout
Definitions
- the present application relates to the field of storage technologies, and in particular, to a method for accessing a storage unit, a method for repairing, a bare chip, and a storage chip.
- the row selection signal selected by the row decoder after performing the row decoding operation on the original row address is the redundant row row select signal.
- the data to be written is stored in the redundant row or the data stored in the redundant row is read.
- the redundant row address is carried in the response message returned to the user. In this process, the user is not aware of the replacement operation of redundant rows.
- the present application provides an access method, a repair method, a bare chip and a memory chip for a storage unit, which can reduce the volume of a product on the basis of realizing failure replacement.
- a method for accessing a storage unit which is applied to a repairing device, and the method includes:
- the access address includes the original address of at least one storage unit in the storage device
- the global repair information includes the original address of the at least one failed unit existing in the storage device and the address used to replace the failed unit.
- the original address in the access address that points to the first failed unit is replaced with the first redundant unit corresponding to the first failed unit redundant address of the unit;
- the replaced access request is sent to the storage device.
- the repair device when the repair device receives an access request including an access address, the repair device identifies whether there is a failed unit in the storage unit pointed to by the access address according to the global repair information, and in the storage unit identified by the access address When there is a first failure unit, replace the original address pointing to the first failure unit in the access address with the redundant address of the redundant unit corresponding to the first failure unit, and send the replaced access request to the storage device, which can realize the utilization of Redundant units repair failed units.
- the storage device is a storage die
- the repair device is a logic die
- the storage device is a storage die
- the repair device is located on a logic die including a controller
- the repair device is connected to the storage die through the controller.
- the storage device is a storage chip
- the repair device is a main chip
- the storage device is a storage chip
- the repair device is located on a main chip including a controller
- the repair device is connected to the storage chip through the controller.
- the memory die may include row decoders. Row decoders are used to interface with the logic die.
- a memory die includes at least one memory cell and at least one redundancy cell.
- the failed unit is a failed storage unit, and the redundant unit is used to replace the failed unit to execute the access request.
- a memory die may contain at least one data storage area and at least one redundancy area.
- Each storage unit in the storage die is located in each data storage area, and each redundant unit is located in each redundant area.
- one data storage area may correspond to one redundant area, and based on this, the redundant unit in the redundant area may only be used to replace the failed unit in the data storage area corresponding to the corresponding redundant area.
- multiple data storage areas may correspond to one redundant area, and based on this, redundant units in the redundant area may be used to replace failed units in multiple data storage areas, that is, multiple data storage areas
- the storage units in a redundant area can share redundant units in a redundant area.
- a memory die may include multiple blocks, and each block may include multiple sectors.
- each block may be provided with a data storage area and a redundancy area.
- each block may be provided with a data storage area, and multiple blocks may be provided with a redundant area.
- each sector may be provided with a data storage area and a redundancy area.
- each sector may be provided with a data storage area, and multiple blocks may be provided with a redundant area.
- each storage unit may be a row located in a data storage area
- each redundant unit may be a row located in a redundant area
- the original address of the storage unit belongs to the original address section, and the redundant address of the redundant unit belongs to the redundant address area; any original address points to any data storage area of the storage die A storage unit, any of the redundant addresses points to a redundant unit in any redundant area in the storage die.
- the method before the receiving the access request including the access address, the method further includes:
- the second failure unit is the same as or different from the first failure unit.
- the repairing device includes a one-time programmable memory efuse
- the method further includes:
- the third failure unit is the same as or different from the first failure unit.
- the method further includes:
- the access response containing the original address of the first failed unit is sent to the bus.
- the storage device is a storage die
- the repair device is located on a logic die including a controller, and the repair device is connected to the storage die through the controller;
- the memory die includes at least two data storage areas, and the global repair information includes the original addresses of all failed cells in the at least two data storage areas and a replacement for each of the at least two data storage areas.
- the redundant address of the redundant unit of the failed unit includes the redundant address of the redundant unit of the failed unit.
- any storage unit is a data row
- any redundant unit is a redundant row
- the first redundant unit and the first failed unit are located in the same sector of the storage device; or, the first redundant unit and the first failed unit different sectors in the same block in the storage device; or, the first redundant unit and the first failure unit are located in different blocks in the storage device.
- a method for accessing a storage unit is provided, applied to a storage device, where the storage device includes at least one storage unit, and the method includes:
- an access request including an access address sent by a repair device, wherein the access address includes a redundant address of a redundant unit in the storage device; wherein the redundant unit is used to replace the at least one storage unit in the the failed unit;
- An access response is sent to the repair device.
- the storage device is a storage die
- the repair device is a logic die
- the storage device is a storage die, the repair device is located on a logic die including a controller, and the repair device is connected to the storage die through the controller; or,
- the storage device is a memory chip, and the repair device is a main chip; or,
- the storage device is a storage chip, the repair device is located on a main chip including a controller, and the repair device is connected to the storage chip through the controller.
- the method before receiving the access request including the access address sent by the repairing apparatus, the method further includes:
- the original address of the failed unit is sent to the repair device, so that the repair device allocates the redundant unit for replacing the failed unit,
- the corresponding relationship between the original address of the failed unit and the redundant address of the redundant unit is added to the global repair information.
- a method for repairing a storage unit is provided, applied to a storage device, and the method includes:
- the original address of the failed unit is sent to the repair device, so that the repair device is a redundant unit allocated to replace the failed unit.
- sending the original address of the failed unit to the repair device includes:
- the original address of the defective unit is sent to the repair device.
- a method for repairing a storage unit which is applied to a repairing device, and the method includes:
- the original address of the failed unit and the redundant address of the redundant unit are added to the global repair information.
- the repairing device includes efuse;
- the receiving, sent by the storage device, the original address of the failed unit existing in the storage unit in the storage device includes: receiving all the data sent by the storage device. the original address of at least one failed unit existing before encapsulation in the storage device;
- the allocating a redundant unit for replacing the failed unit includes: allocating a redundant unit for replacing each of the failed units existing before encapsulation;
- the adding the original address of the failed unit and the redundant address of the redundant unit to the global repair information includes: adding the original address of each failed unit existing before encapsulation and the redundancy corresponding to each failed unit The correspondence of redundant addresses of units is stored in the efuse; after each power-on of the repair device, the original address of at least one failed unit existing in the storage device read from the efuse The corresponding relationship with the redundant addresses of the redundant units corresponding to each failed unit is added to the global repair information.
- a repair device comprising:
- the transceiver module is configured to receive an access request including an access address; the access address includes an original address of at least one storage unit in the storage device.
- the processing module is configured to identify whether there is a failed unit in the storage unit pointed to by the access address according to the global repair information; wherein, the global repair information includes the original address and usage of at least one failed unit existing in the storage die. the redundant address of the redundant unit to replace each of the failed units in the at least one failed unit; when it is recognized that there is a first failed unit in the storage unit pointed to by the access address, point the access address to the redundant unit The original address of the first failed unit is replaced with the redundant address of the first redundant unit corresponding to the first failed unit.
- the transceiver module is further configured to send the replaced access request to the storage device.
- the transceiver module is further configured to receive, before the receiving the access request including the access address, the original address of the second failure unit existing in the storage device sent by the storage device;
- the processing module is used for assigning a second redundant unit for replacing the second failed unit; and for adding the corresponding relationship between the original address of the second failed unit and the redundant address of the second redundant unit to the global repair information; wherein, the second failure unit and the first failure unit may be the same or different.
- the repair device includes a one-time programmable memory efuse; the transceiver module is further configured to receive the storage device sent by the storage device before receiving the access request including the access address.
- the corresponding relationship between the original addresses of the three failed units and the redundant addresses of the third redundant unit is stored in the efuse; and is used for, after each power-on of the repair device, read from the efuse
- the corresponding relationship between the original address of the third failed unit and the redundant address of the third redundant unit existing in the storage device is added to the global repair information.
- the transceiver module is further configured to receive an access response including the redundant address of the first redundant unit sent by the storage device; the processing module is further configured to obtain the the original address of the first failed unit corresponding to the first redundant unit; adding the original address of the first failed unit to the access response; the transceiver module is further configured to include the first failed unit The original address of the access response is sent to the bus
- a storage device comprising:
- a transceiver module configured to receive an access request including an access address sent by a repair device, wherein the access address includes a redundant address of a redundant unit in the storage device; wherein the redundant unit is used to replace the a failed unit in at least one storage unit;
- the processing module is configured to access the redundant address; the transceiver module is further configured to send an access response to the repair device.
- the processing module is further configured to detect whether there is a failed unit in the storage unit in the storage device before receiving the access request including the access address sent by the repairing device; the transceiver module is further configured to , when there is a failed unit in the storage unit in the storage device, send the original address of the failed unit to the repair device, so that the repair device allocates the redundant unit for replacing the failed unit , and the corresponding relationship between the original address of the failed unit and the redundant address of the redundant unit is added to the global repair information.
- a storage device comprising:
- the processing module is used for detecting whether there is a failed unit in the storage device.
- the transceiver module is configured to send the original address of the failed unit to the repairing device when there is a failed unit in the storage unit in the storage device.
- the processing module is specifically used to detect whether there is a failed unit in the storage unit in the storage device before the storage device is packaged; the transceiver module is specifically used for the storage unit in the storage device. When there is a failed unit in the storage unit, the original address of the failed unit is sent to the repairing device.
- a repair device comprising:
- the transceiver module is configured to receive the original address of the failed unit existing in at least one storage unit of the storage device sent by the storage device.
- a processing module configured to allocate a redundant unit for replacing the failed unit; and configured to add the original address of the failed unit and the redundant address of the redundant unit to the global repair information.
- the repairing device includes a one-time programmable memory efuse
- the transceiver module is specifically configured to receive the original address of at least one failed unit existing in the storage device before encapsulation sent by the storage device; the processing module is specifically configured to allocate an redundant unit; and be specifically used for storing the correspondence relationship between the original address of each failed unit before encapsulation and the redundant address of the redundant unit corresponding to each failed unit in the efuse; in the repair device After each power-on, the corresponding relationship between the original address of at least one failed unit existing in the storage device read from the efuse and the redundant address of the redundant unit corresponding to each failed unit is added to the corresponding relationship. global repair information.
- a repair device comprising: an interface, a processor and a memory;
- the interface is used to connect a storage device and a bus; the memory is used to store instructions; the processor is used to execute the instructions to implement any one of the first aspect and the fourth aspect or any possible implementation manner. method.
- the storage device is a storage die
- the repair device is a logic die
- the storage device is a storage die, the repair device is located on a logic die including a controller, and the repair device is connected to the storage die through the controller; or,
- the storage device is a memory chip, and the repair device is a main chip; or,
- the storage device is a storage chip, the repair device is located on a main chip including a controller, and the repair device is connected to the storage chip through the controller.
- the memory further includes: a one-time programmable memory efuse.
- the processor may include: a DRAM controller.
- a storage device comprising: an interface, a processor and a memory;
- the interface is used to connect the repair device and the bus;
- the memory is used to store instructions and data;
- the processor is used to execute the instructions to implement any one of the second aspect or the third aspect or any possible implementation manner method in .
- the processor may include: a row address decoder.
- each storage unit includes at least one data row
- each redundant unit includes at least one redundant row
- a memory chip including:
- the logic die includes a repair device and a controller, the repair device communicates with the storage die through the controller;
- the logic die is used to perform the method in any one of the first aspect or the fourth aspect or any possible implementation;
- the memory die is used to perform the method of any of the second or third aspects or any possible implementation.
- a twelfth aspect provides a memory chip, comprising:
- the logic die is used for executing the method in any possible implementation manner of the first aspect, and the storage die is used for executing the method in any possible implementation manner in the second aspect; or,
- the logic die is used for executing the method in any possible implementation of the fourth aspect, and the memory die is used for executing the method in any possible implementation in the third aspect.
- an apparatus in yet another aspect, includes a processing module and a transceiver module, and the processing unit executes an instruction to control the apparatus to execute the method in the first aspect or any possible design of the first aspect.
- the apparatus may further include a storage module.
- the device may be a memory chip.
- the processing module may be a processor, and the transceiver module may be a transceiver; if a storage module is further included, the storage module may be a memory.
- the processing module may be a processor, and the transceiver module may be an input/output interface, a pin or a circuit, etc.; if a memory module is also included, the memory module may be an in-chip memory module.
- the storage module (eg, register, cache, etc.) of the chip can also be a storage module outside the chip (eg, read-only memory, random access memory, etc.).
- the processor mentioned in any of the above may be a general-purpose central processing unit (Central Processing Unit, referred to as CPU), a microprocessor, a specific application integrated circuit (application-specific integrated circuit, referred to as ASIC), or an or A plurality of integrated circuits for controlling program execution of the spatial multiplexing method of the above aspects.
- CPU Central Processing Unit
- ASIC application-specific integrated circuit
- a computer-readable storage medium having stored therein instructions executable by one or more processors on a processing circuit. When run on a computer, the computer is caused to perform the method of the first aspect above or any possible implementation thereof.
- a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of the first aspect above or any possible implementation thereof.
- FIG. 1 is a schematic structural diagram of a memory chip in an embodiment of the present application.
- FIG. 2 is a schematic diagram of a layered structure of a die in a memory chip according to an embodiment of the present application
- FIG. 3 is a schematic diagram of a logical structure of a memory chip in an embodiment of the present application.
- FIG. 4 is a schematic diagram 1 of an interaction flow of a method for accessing a storage unit provided by an embodiment of the present application
- FIG. 5 is a schematic diagram of an interaction flow of a method for repairing a storage unit according to an embodiment of the present application
- FIG. 6 is a schematic diagram of a system architecture and a processing process when the method for accessing a storage unit provided by an embodiment of the present application is applied to a DRAM;
- FIG. 7 is a schematic structural diagram 1 of a repairing device provided by an embodiment of the present application.
- FIG. 8 is a schematic structural diagram 1 of a storage device provided by an embodiment of the present application.
- FIG. 9 is a second structural schematic diagram of a repairing device provided by an embodiment of the present application.
- FIG. 10 is a second schematic structural diagram of a storage device provided by an embodiment of the present application.
- the embodiments of the present application provide a storage unit access method, a repair method, and related technical solutions.
- a chip with a memory function may be referred to as a memory chip.
- the memory chip can be dynamic random access memory (DRAM), static random access memory (SRAM), non-volatile memory (Non-Volatile Memory, NVM), etc. .
- DRAM dynamic random access memory
- SRAM static random access memory
- NVM non-volatile memory
- the memory chip may be composed of one or more dies.
- the manufacturer of the storage chip needs to perform quality inspection on the bare chips that make up the chip, and after the inspection, package the multiple bare chips that make up the chip to obtain the storage chip.
- FIG. 1 is a schematic structural diagram of a memory chip in an embodiment of the present application.
- the memory chip 10 may include one logic die 100 and at least one storage die 200 .
- one or more storage particles can be deployed on each storage die.
- the storage granules deployed on each storage die may be divided into one or more blocks (banks), and each bank may be divided into one or more sectors (sectors). Each sector can be divided into multiple rows (row).
- Each memory die may also be deployed with an address decoder, such as a row address decoder, which may also be referred to as a row decoder.
- the memory chip may further include an outer packaging layer.
- FIG. 2 is a schematic diagram of a layered structure of a die in a memory chip according to an embodiment of the present application.
- the logic die and the storage die in the memory chip may be arranged in layers in a 3D stacking manner, wherein the logic die is located on the bottom layer, and the storage die is located above the logic die , the projection of the storage die on the plane where the logical die is located can occupy the same area as the logical die.
- the multiple storage die is also hierarchically set, and each storage die is placed on the logical die. The area occupied by the projection of the plane on which the slices lie can be the same.
- one memory chip may include one logic die and an integer multiple of four memory die.
- the logic die may also be referred to as a base film, or a base die.
- FIG. 3 is a schematic diagram of a logical structure of a memory chip in an embodiment of the present application.
- the logic die 100 may include: a repair apparatus 1001 and a controller 1002 .
- the logic die may also include an external interface to communicate with the system bus, and an internal interface to communicate with the various memory dies.
- the memory chip may also communicate with the system bus through the main chip, wherein the main chip may be provided with a repair device and a controller, and the controller may communicate with the system bus through the main chip.
- the logic die accesses the memory cells in each memory die on the memory chip, and the repair device and the controller in the main chip can perform a method similar to the repair device and controller in the structure shown in FIG.
- the controller and the repair device may not be provided in the logic die of the memory chip in the MCU.
- the chip scale is getting larger and the operating frequency is getting higher and higher, for example, DRAM is a large-capacity and high-density semiconductor memory.
- DRAM is a large-capacity and high-density semiconductor memory.
- redundant rows may be deployed in the storage die of the memory chip to replace the failed data rows.
- the repair technology based on redundant rows may be referred to as a row redundancy (Row Redundancy) technology .
- the storage die may include: at least one storage unit and at least one redundant unit.
- a storage die may be divided into rows on the grains deployed on itself, each row comprising one or more smallest cells.
- the storage die can divide multiple rows divided on its own grain into at least one data storage area and at least one redundant area, and configure the original address segment for each data storage area, and configure redundant addresses for each redundant area segment, wherein each original address segment and each redundant address segment do not overlap each other.
- each data storage area may include at least one storage unit, each storage unit may include one data row or multiple data rows, and the original address of each storage unit belongs to the data storage area to which the storage unit belongs.
- Original address segment; each redundant area may include at least one redundant unit, each redundant unit may include 1 redundant row or multiple redundant rows, and the redundant address of the redundant unit of each redundant area It belongs to the redundant address segment corresponding to the redundant area to which the redundant unit belongs.
- the number of rows contained in each redundant cell may be the same as the number of rows contained in each memory cell.
- the failed memory cell can be called a failed cell, and the memory chip uses the redundant cell located in the redundant area to replace the failed cell to complete the data access operation to improve the chip yield.
- M data storage areas and N redundancy areas may be deployed on the storage die.
- M and N are positive integers, and in one example, M ⁇ N.
- the correspondence between the M data storage areas and the N redundant areas can be flexibly configured.
- one data storage area may correspond to one redundant area, and based on this, the redundant unit in the redundant area may only be used to replace the failed unit in the data storage area corresponding to the corresponding redundant area.
- multiple data storage areas may correspond to one redundant area, and based on this, redundant units in the redundant area may be used to replace failed units in multiple data storage areas, that is, multiple data storage areas The storage units in a redundant area can share redundant units in a redundant area.
- a memory die may include multiple banks, and each block may include multiple sectors.
- each bank can have a data storage area and a redundancy area.
- each bank can set up a data storage area, and multiple banks can set up a redundant area.
- each sector can have a data storage area and a redundancy area.
- each sector can set up a data storage area, and multiple sectors can set up a redundant area.
- each storage unit may include at least one data row, and the number of redundant rows included in each redundant unit may be the same as the number of data rows included in each storage unit.
- each sector set a data storage area and a redundant area as an example, the ratio of the number of rows located in the data storage area to the number of redundant rows located in the redundant area in each sector can be agreed Scale, for example 2K:16.
- each storage die contains multiple banks, wherein each sector in a part of the bank may deploy a redundant area, and each sector in another part of the bank may not deploy a redundant area.
- a bank may contain multiple sectors, wherein some sectors may deploy redundant areas, and other sectors may not deploy redundant areas. The embodiments of the present application are not limited thereto.
- FIG. 4 is a schematic diagram 1 of an interaction flow of a method for accessing a storage unit according to an embodiment of the present application.
- the execution body of the embodiment of the present application may include: a repair device and a storage device.
- the storage device may be a storage die including at least one storage unit and at least one redundant unit.
- the repair device may be located on a logic die containing a controller, and the repair device may communicate with the storage die through the controller.
- the repair apparatus receives an access request including an access address from a bus; the access address includes an original address of at least one storage unit in the storage apparatus.
- the repairing device may receive an access request from an IO bus through an external interface.
- the access request can be a read command or a write command.
- the read instruction may contain the memory address of the memory cell requested to be read.
- the write instruction may include a request for data to be written, and a target address of the data, that is, a storage address of a storage unit where the data is to be stored.
- the access request may also be a refresh request.
- the access request may be any data processing request including a storage address of data.
- the storage device may configure an original address segment for all storage units in the storage device, and configure all redundant storage units in the storage device with an original address segment.
- the unit configures redundant address segments.
- the storage die includes at least one data storage area and at least one redundant area, each data storage area includes at least one storage unit, and the original address of each storage unit belongs to the original address segment; each redundant area At least one redundant unit is included, and the redundant address of each redundant unit belongs to the redundant address segment.
- the memory chip only presents the original address segment to the outside world, that is, the address included in the access request is the original address of the storage unit.
- the access address may be a logical block address (Logical Block Address, LBA), also referred to as a logical block address.
- LBA Logical Block Address
- the original address area is LBA001 to LBA399
- the redundant address area may be LBA400 to LBA420.
- the repair device identifies whether there is a failed unit in the storage unit pointed to by the access address according to the global repair information; wherein the global repair information includes the original address of at least one failed unit existing in the storage device and the replacement A redundant address of a redundant unit of each of the at least one failed unit.
- the repair apparatus may identify whether the access address includes the original address of the failed unit according to the global repair information.
- the global repair information may be stored on the repair device, and the repair device may continuously update the global repair information during the operation of the chip.
- the repair device can learn the information of the failed unit detected by the storage device during the process of reading and writing data, refreshing data, detecting and maintaining, etc., and assigns redundant units to it, and adds newly detected units to the global repair information.
- the global repair information may include original addresses of all detected failed cells in the storage device and redundant addresses of redundant cells used to replace each failed cell.
- the repairing device can also detect the original address of the failed unit by the storage device in the testing phase and the redundant unit allocated by the repairing device for each failed unit.
- the corresponding relationship information of the redundant addresses is stored in the efuse.
- the repairable device can read the repair information recorded in the test phase from the efuse as the initial global repair information.
- the repair device when recognizing that there is a first failed unit in the storage unit pointed to by the access address, the repair device replaces the original address pointing to the first failed unit in the access address with the redundant address of the first redundant unit corresponding to the first failed unit .
- the access address contains an original address pointing to the first failed unit
- the first failure unit is any failure unit.
- the access address includes the first original address
- the logic die queries the global information table to determine that the first storage unit pointed to by the first original address has failed, and the failed first storage unit is the first failed unit, and further The global information table is queried to obtain the first redundant address corresponding to the first original address, wherein the first redundant unit pointed to by the first redundant address is used to replace the first failed unit pointed to by the first original address.
- the access address may include the address range LBA001 to LBA010, assuming that LBA002 and LBA004 are the original addresses of the failed units recorded in the global repair information, if the original address is the redundant address of the redundant unit corresponding to the storage unit of LBA002 is LBA500, the redundant address of the redundant unit corresponding to the storage unit whose original address is LBA004 is LBA501, then the replaceable access addresses can be LBA001, LBA500, LBA003, LBA501, LBA005 to LBA010.
- the repairing device sends the replaced access request to the storage device.
- the replaced access request includes the first redundant address corresponding to the first redundant unit.
- the storage device is a storage die
- the repair device located on the logical die can send the replaced access request to the storage die through the controller on the logical die.
- the storage device completes the access process based on the redundant unit pointed to by the redundant address.
- the storage device can access redundant addresses.
- the access request when the access request is a write command, the access request further includes data to be written, the storage device can write the data to be written into the redundant unit pointed to by the redundant address, and then the storage device can generate an access response. , which contains the redundant addresses of the redundant cells that store the data to be written.
- the storage device when the access request is a read command, the storage device may read data stored in the redundant unit pointed to by the redundant address, including data read from the redundant unit pointed to by the redundant address.
- the storage device sends an access response to the repair device, where the access response includes the redundant address of the first redundant unit.
- the access response when the access request is a write command, the access response includes the redundant address of the redundant unit storing the data to be written. In yet another example, when the access request is a read command, the access response includes data read from the redundant unit pointed to by the redundant address.
- the storage device is a storage die
- a repair device located on the logical die may receive an access response sent by the storage device through a controller on the logical die.
- the repairing device obtains, according to the global repair information, the original address of the first failed unit corresponding to the redundant address of the first redundant unit, and replaces the redundant address pointing to the first redundant unit in the access response is the original address of the first failed unit.
- the access response may contain an address range consisting of LBA001, LBA500, LBA003, LBA501, LBA005 to LBA010, assuming that LBA500 and LBA501 are the redundant addresses of the redundant units recorded in the global repair information, if the redundant addresses are The original address of the storage unit corresponding to the storage unit of LBA500 is LBA002, and the original address of the failed unit corresponding to the redundant unit whose redundant address is LBA501 is LBA004, so the replaceable access addresses can be LBA001 to LBA010.
- the global repair information may support using the original address of the failed unit to query the redundant address of the corresponding redundant unit, and may also support using the redundant address of the redundant unit to reversely query the original address of the corresponding failed unit.
- the repair apparatus may store the global repair information as redundant address mapping information and original address reverse mapping information, respectively, and the repair apparatus may query the original address of the storage unit according to the redundant address mapping information when receiving an access request. The address corresponds to the redundant address of the redundant unit. When receiving the access response, the repairing apparatus may also query the original address of the corresponding storage unit according to the redundant address of the redundant unit.
- the repair device sends the replaced access response to the bus.
- the restoration device and the storage device cooperate to realize the process of completing data access based on the redundant unit.
- the restoration device realizes the redundancy replacement and restoration according to the global restoration information, and the control function of the redundancy replacement restoration can be implemented on the logic die with more advanced technology, and the storage die does not need to be provided with support A circuit for switching row selection signals, wherein the circuit for switching row selection signals is used to directly transfer the row selection signal of the failed row pointing to the normal data storage area to the row in the redundant area for replacing the redundant row of the failed row circuit for marking signals.
- the circuit structure on the memory die is greatly simplified.
- the technical solutions provided by the embodiments of the present application can reduce the area of the tiled area occupied by the storage die by simplifying unnecessary circuits in the storage die without reducing the storage capacity.
- the area of the tiled area occupied by the final memory chip product can be greatly reduced, that is, with a smaller die.
- the area realizes the redundant replacement function, saves the storage die area, and makes the product more miniaturized.
- each redundant unit may be one row in the sector.
- the minimum replacement unit can realize more fine-grained planning and utilization of redundant resources, and flexibly and comprehensively repair the failed unit in each sector. For example, when there are 3 failed rows in a sector, and the interval between the 3 failed rows exceeds 8 rows, if the minimum replacement unit is 8, the 16 redundant row resources can actually only repair 2 failed rows. Then the sector cannot be repaired.
- the minimum replacement unit is one row, 16 redundant row resources can repair the three failed rows, so that the sector can be repaired and used.
- Embodiments of the present application further provide technical solutions related to a method for repairing a group of storage units.
- the method for repairing provided by the embodiments of the present application may be executed before executing the method for accessing storage units in the foregoing embodiments.
- the repairing apparatus may acquire or update the global repairing information through the repairing method for the storage unit provided in the embodiment of the present application.
- the repair apparatus may update the global repair information in the following manner.
- the embodiment of the present application may further include:
- the storage device detects whether there is a failed unit in the storage units in the storage device.
- the storage device may be a storage die.
- the memory die may detect a failed unit when executing instructions such as a read data command, a data write command, and a refresh command. This embodiment of the present application does not limit this.
- the failure unit detected by the storage device may be the second failure unit, where the second failure unit may be the first failure unit in the foregoing embodiment or another failure unit.
- the storage device sends the detected original address of the failed unit to the repairing device.
- the storage device is a storage die, and when the repair device is located on a logic die including the controller, the storage device may send the original data of the failed unit to the repair device through the controller. As an example, the storage device may send the original address of the failed unit to the repair device when there is a failed unit.
- the original address of the detected failed unit may be recorded as the original address of the second failed unit.
- the repairing apparatus allocates a redundant unit for replacing the detected failed unit.
- the redundant unit used to replace the second failed unit may be the second redundant unit.
- the storage device may include at least one data storage area and at least one redundant area.
- each data storage area includes at least one storage unit
- each redundant area includes at least one redundant unit
- the storage device can allocate original address segments to the storage units belonging to the respective data storage areas, and, for the storage units belonging to the respective redundant areas
- the redundant units of the area are allocated redundant address segments.
- the addresses of any redundant address segment and any original address segment do not overlap.
- the storage device may include M data storage areas and N redundancy areas, where M and N are positive integers. As an example, M may be greater than or equal to N.
- redundant units may be allocated based on the principle of proximity.
- the storage device may be provided with X functional areas, wherein each functional area may include 1 data storage area, and the number of redundant areas in at least one of the X functional areas is not zero.
- each functional area may include 1 data storage area, and the number of redundant areas in at least one of the X functional areas is not zero.
- the repairing device allocates redundant units, it can preferentially allocate redundant units belonging to the same functional area to the failed units. If the redundant units belonging to the same functional area have been allocated, redundant units in other functional areas can be selected.
- the functional area may be a sector or a block.
- the storage device may further set a high-level functional area and a low-level functional area, wherein each high-level functional area includes at least one low-level functional area, each low-level functional area includes at least one data storage area, and all of the low-level functional areas include at least one low-level functional area.
- the number of redundant areas of at least one lower-level functional area is not zero.
- high-level functional areas can be blocks and low-level functional areas can be sectors. Each block may include at least one sector.
- the storage device may include block 0 and block 1, block 0 includes sector 0 (denoted as S01) and sector 1 (denoted as S02), and block 1 includes sector 0 (denoted as S11).
- the storage device may set a redundant area on at least one of the four sectors S01, S02, S11 and S12.
- the storage device may include a first storage unit located at S01 and a first redundant unit located at S01, and in yet another example, the storage device may include a first storage unit located at S01 and a first redundant unit located at S02. In yet another example, the storage device may include a first storage unit located at S01 and a first redundant unit located at S11.
- the repair device can preferentially select the redundant unit in the redundant area corresponding to the same low-level functional area as the first redundant unit; if the redundant units in the redundant area corresponding to the same low-level functional area have been allocated, the same In the high-level functional area, select the redundant unit in the redundant area of the nearest low-level functional area; if the redundant units in the redundant area corresponding to the nearest low-level functional area have been allocated, the same high-level functional area can be used. Redundant units in the redundant areas of other low-level functional areas in the area; if the redundant units in the redundant areas of the same high-level functional area have been allocated, you can select the redundant units in the redundant areas of other high-level functional areas. redundant unit.
- the redundant unit in is the first redundant unit corresponding to the first failed unit.
- the redundant units in the redundant area belonging to the same sector have been allocated, select the redundant unit in the redundant area in the nearest sector that belongs to the same bank as the first data storage area as the first data storage area.
- the first redundant unit corresponding to the failed unit is selected as the corresponding first failed unit.
- the redundant cells in the redundant areas in all banks on the storage die select the redundant cells in the redundant area in the storage die closest to the storage die as the redundant cells in the redundant area.
- the first redundant unit corresponding to the first failed unit may be located in the same sector, or, alternatively, the first failed unit and the first redundant unit may be located in different sectors of the same bank , or, the first failed unit and the first redundant unit may be located in different banks of the same memory die.
- the embodiment of the present application can actually realize cross-sector/cross-block sharing of global redundant resources, effectively improving the utilization rate of redundant resources , thereby improving the chip yield.
- the failure unit of each sector and each block can be repaired. For example, when the number of rows of failed units in a sector exceeds the number of rows of redundant units configured in the sector, a comprehensive repair can be performed by scheduling idle redundant units in other sectors to avoid appearing in the entire storage system. When there are spare redundant units on the chip, it is still impossible to repair the situation that a large number of failed units in a certain sector are concentrated.
- the repair apparatus adds the original address of the detected failure unit and the allocated redundant address of the redundant unit to the global repair information.
- the global repair information may include one or more groups of corresponding relationships between the failed units and the redundant units.
- steps S201 to S204 may also be executed after step S101.
- this repair method when the storage device detects a failed unit, it can notify the repair device to allocate the corresponding redundant unit, so that the global update can be dynamically updated during the process of executing read and write instructions, refresh instructions, etc. that can trigger the detection of the failed unit. Repair information.
- the access method of the storage unit is executed, the read and write operations can be performed on the redundant unit, and the use of the failed unit can be avoided, so that the access speed of the storage chip can be improved.
- FIG. 5 is a schematic diagram of an interaction flowchart of a method for repairing a storage unit provided by an embodiment of the present application.
- the repairing apparatus may also obtain the global repairing information in the following manner.
- a one-time programmable memory may also be deployed in the repair device or the logic die where the repair device is located.
- the corresponding information of the replacement unit that is, the repair information recorded in the test phase, is stored in efuse.
- the repair device can read the repair information written in the test phase from the efuse as the initial global repair information, and continuously update the global repair information in combination with the method in FIG. 4 .
- the steps in this embodiment of the present application may further include:
- the repair device sends a failure detection indication to the storage device.
- the failure detection indication may be a test command obtained from the IO bus, for example, may be a write command, a read command or a refresh command, wherein the carrying access address is the original address corresponding to all rows on the storage die.
- the repair device may send various messages to the storage device through the controller.
- the storage device detects whether there is a failed unit in the storage unit.
- the failure unit detected by the presence device before packaging may be the third failure unit, and the third failure unit may be the first failure unit in the foregoing embodiment, or may be another failure unit.
- the storage device sends the detected original address of the failed unit to the repairing device.
- the repair apparatus allocates a redundant unit for replacing the failed unit.
- the redundant unit used to replace the third failed unit may be denoted as the third redundant unit.
- steps S302-S304 is similar to that of S201-S203, and reference may be made to the relevant descriptions in S201-S203.
- the repairing device writes the original addresses of all the failed units and the redundant addresses of the corresponding redundant units into the efuse.
- the original addresses of all the failed units and the redundant addresses of the corresponding redundant units recorded in the test phase may be referred to as test repair information.
- the repairing apparatus writes the original addresses of all the failed cells detected before encapsulation and the redundant addresses of the redundant cells corresponding to each failed cell into the efuse at one time.
- the repair device when powered on, the repair device reads the original addresses of all failed units and the redundant addresses of the corresponding redundant units stored in the efuse, and uses them as initial global repair information.
- the repair device can directly obtain this part of the repair information every time it is powered on, which avoids repeated detection after each power-on, thereby reducing the need for repeated detection and repair work.
- the energy consumption of the bare chip improves the access efficiency of the memory chip.
- the above two repair methods can respectively perform the repair work of the memory cell when the memory chip is in the working state and when the memory chip is in the test stage, and the memory chip can support the execution of the above two repair methods at the same time.
- FIG. 6 is a schematic diagram of a system architecture and a processing flow when the method for accessing a memory cell provided by an embodiment of the present application is applied to a DRAM.
- the memory chip in this embodiment of the present application may include: a logic die 610 and a DRAM die (DRAM die) 620 .
- the logic die may include: a DRAM controller (DRAM controller) 601, a redundant row address mapping (Redundent Row Address Mapping) module 603, a global repair information (Repair info) management module 604, an original address reverse mapping (Original Row Address Reverse Mapping) module 605.
- the modules 603 to 605 may constitute the repair device 602 .
- the DRAM die 620 may include a row decoder (Row Decoder) 606 and a memory array arranged as blocks and sectors (the right portion of the row decoder in FIG. 6).
- the row decoder is also called a row address decoder or a row address decoding module.
- redundant storage resources used for failure repair can be evenly distributed in each Sector or Bank on the DRAM Die , for example, every 2K rows can be equipped with 16 redundant rows (as shown by the gray block in Figure 6); on the other hand, the control of replacing and repairing failed rows with redundant rows is implemented on the logic die, and on the DRAM die It is no longer necessary to set up replacement repair related logic.
- the storage unit is one data row and the redundant unit is one redundant row, wherein each row can be recorded as 1WLs).
- the steps in this embodiment of the present application may include two parts: steps in the chip testing process and steps in the normal use phase of the chip.
- the steps in the chip testing stage may include:
- Step 1 If a DRAM bare chip finds a failed cell in the normal data storage area, it is determined which redundant row to use for replacement according to the allocation strategy. For example, according to the principle of proximity, the failed cell located in the same sector or the same bank can be preferentially selected. the failed unit. All the repair information of the memory chip is burned and written on the logic die, and it needs to be recorded in the global repair information management module by programming efuse. In the normal use scenario, when the chip is powered on, the global repair information management module reads the repair information recorded in the test phase from the efuse and transfers it to the redundant row address mapping module and the original row address reverse mapping module.
- the global repair information management module on the logic die is mainly composed of efuse resources, and is used to record the global redundant row repair information of the chip.
- the failure repair information can record the target failure row that needs to be replaced according to the redundant behavior unit, and also need to consider mechanisms such as fault tolerance and supplementary records.
- the global repair information management module can also be called the efuse module.
- Step 2 in the normal use scenario, when the user initiates read and write access to the chip, the read and write commands will first enter the redundant row address mapping module for processing, and the redundant row address mapping module identifies the invalid row in the original access address. address, and replace the original invalid row address with the redundant row address according to the determined replacement strategy.
- a content addressable memory (Content Addressable Memory, CAM) is used in the redundant row address mapping module to form a failed row address mapping table with redundant row addresses.
- the CAM search is performed for the row address in each access instruction, and the invalid row address in the match will be replaced with the corresponding CAM address (ie, redundant row address).
- the replacement strategy may be the principle of proximity, the principle of balance, etc. For example, the redundant unit in the redundant area in the sector with the lowest usage rate in the same bank is preferentially selected.
- Step 3 the access command after the invalid row address is replaced by the redundant row address will normally enter the controller for processing.
- the controller sends corresponding commands to the DRAM die according to the DRAM timing.
- Step 4 The row address decoding module on the DRAM bare chip decodes the received row address into a row selection signal, among which redundant rows are also addressed uniformly, and the addressing of redundant rows can be within the address range of normal data rows. outside). For the instruction that has been replaced with the redundant row address, it is directly decoded as the row select signal of the redundant row. At this time, all access operations are actually operations on the smallest unit in the redundant row.
- each storage unit may include at least one row and at least one column of the smallest unit.
- the storage unit includes 1 row and column address.
- Step 5 After the controller completes the operation on the DRAM bare chip, it will return corresponding read and write responses and read data. For read and write operations of redundant rows, the controller will normally return the corresponding redundant row address information to the original address reverse mapping module.
- Step 6 the original address reverse mapping module will check out the redundant row address, and look up an original row address reverse mapping table, after converting the redundant row address in the read and write response to the original row address, the read and write response and The read data is returned to the user side. The user is not aware of the redundant replacement operation of the invalid address.
- the original row address reverse mapping table in the foregoing embodiment may be a global information table.
- the original address reverse mapping module may use the SRAM to form an original row address reverse mapping table with redundant row addresses.
- the address of each failed cell is recorded in the SRAM.
- the redundant row address is used to find the corresponding original address in the SRAM, and reverse replacement is performed .
- the global repair information management module, the redundant row address mapping module, and the original row address reverse mapping module cooperate with other modules to support the global redundancy replacement and repair of the entire chip, and support the fine-grained single row Replacement fixes.
- the embodiment of the present application also provides a repairing device.
- FIG. 7 is a first structural schematic diagram of a repairing device provided by an embodiment of the present application.
- the repairing apparatus in this embodiment of the present application may include: a transceiver module 801 and a processing module 802 , wherein the transceiver module is configured to communicate with a bus and a storage device.
- the repair apparatus may further include a storage module 803 .
- the transceiver module is configured to receive an access request including an access address; the access address includes an original address of at least one storage unit in the storage device.
- a processing module configured to identify whether there is a failed unit in the storage unit pointed to by the access address according to the global repair information; wherein, the global repair information includes the original address of at least one failed unit existing in the storage device and an A redundant address of a redundant unit that replaces each of the failed units in the at least one failed unit; when it is recognized that there is a first failed unit in the storage unit pointed to by the access address, pointing the access address to the The original address of the first failed unit is replaced with the redundant address of the first redundant unit corresponding to the first failed unit.
- the transceiver module is further configured to send the replaced access request to the storage device.
- the transceiver module is further configured to receive, before the receiving the access request including the access address, the original address of the second failure unit existing in the storage device sent by the storage device;
- the processing module is used for assigning a second redundant unit for replacing the second failed unit; and for adding the corresponding relationship between the original address of the second failed unit and the redundant address of the second redundant unit to the global repair information, wherein the second failure unit and the first failure unit are the same or different.
- the repair device includes a one-time programmable memory efuse; the transceiver module is further configured to receive the storage device sent by the storage device before receiving the access request including the access address. the original address of the third failed unit existing before encapsulation in the device; the processing module is further configured to allocate a third redundant unit used to replace the third failed unit existing before encapsulation; The corresponding relationship between the original address of the third failed unit and the redundant address of the third redundant unit existing before encapsulation is stored in the efuse; The corresponding relationship between the original address of the third failed unit and the redundant address of the third redundant unit in the storage device read in efuse is added to the global repair information; wherein, the third failed unit and the third A failed unit is the same or different.
- the transceiver module is further configured to receive an access response including the redundant address of the first redundant unit sent by the storage device; the processing module is further configured to obtain the the original address of the first failed unit corresponding to the first redundant unit; adding the original address of the first failed unit to the access response; the transceiver module is further configured to include the first failed unit The original address of the access response is sent to the bus.
- the transceiver module is configured to receive the original address of the failed unit existing in at least one storage unit of the storage device sent by the storage device.
- a processing module configured to allocate a redundant unit for replacing the failed unit; and configured to add the original address of the failed unit and the redundant address of the redundant unit to the global repair information.
- the repairing device includes a one-time programmable memory efuse
- the transceiver module is specifically configured to receive the original address of at least one failed unit existing before encapsulation in the storage device sent by the storage device; redundant unit; and be specifically used for storing the correspondence relationship between the original address of each failed unit before encapsulation and the redundant address of the redundant unit corresponding to each failed unit in the efuse; in the repair device After each power-on, the corresponding relationship between the original address of at least one failed unit existing in the storage device read from the efuse and the redundant address of the redundant unit corresponding to each failed unit is added to the corresponding relationship. global repair information.
- Embodiments of the present application further provide a storage device.
- FIG. 8 is a first schematic structural diagram of a storage device according to an embodiment of the present application.
- the storage device in this embodiment of the present application may include: a transceiver module 901 and a processing module 902 , wherein the transceiver module is configured to communicate with the bus and the storage device.
- the storage device may further include a storage module 903 .
- a transceiver module configured to receive an access request including an access address sent by a repair device, wherein the access address includes a redundant address of a redundant unit in the storage device; wherein the redundant unit is used to replace the a failure unit in at least one storage unit; a processing module for accessing the redundant address; a transceiver module for sending an access response to the repairing device.
- the processing module is further configured to detect whether there is a failed unit in the storage unit in the storage device before receiving the access request including the access address sent by the repairing device; the transceiver module is further configured to , when there is a failed unit in the storage unit in the storage device, send the original address of the failed unit to the repair device, so that the repair device allocates the redundant unit for replacing the failed unit , and the corresponding relationship between the original address of the failed unit and the redundant address of the redundant unit is added to the global repair information.
- the processing module is used for detecting whether there is a failed unit in the storage device.
- the transceiver module is configured to send the original address of the failed unit to the repair device when there is a failed unit in the storage unit in the storage device.
- the processing module is specifically used to detect whether there is a failed unit in the storage unit in the storage device before the storage device is packaged; the transceiver module is specifically used for the storage unit in the storage device. When there is a failed unit in the storage unit, the original address of the failed unit is sent to the repairing device.
- FIG. 9 is a first schematic structural diagram of a repairing apparatus provided by an embodiment of the present application; as shown in FIG. 9 , the repairing apparatus in this embodiment of the present application may include: an interface 1110, a processor 1120, and a memory 1120; wherein, the interface is used for A storage device is connected; the memory is used to store the instructions; the processor is used to execute the instructions to implement the methods in the foregoing embodiments.
- the repair apparatus may also include a bus 1160 for internal interconnection.
- the storage device is a storage die, and the repair device is a logic die; or, the storage device is a storage die, and the repair device is located on a logic die including a controller , and the repair device is connected to the storage die through the controller; or, the storage device is a memory chip, and the repair device is a main chip; or, the storage device is a memory chip, and the repair device A device is located on a host chip that includes a controller, and the repair device is connected to the memory chip through the controller.
- the memory further includes: a one-time programmable memory efuse.
- the memory die may be a DRAM die and the controller may be a DRAM controller.
- FIG. 10 is a schematic structural diagram 1 of a storage device provided by an embodiment of the present application; as shown in FIG. 10 , the storage device in the embodiment of the present application may include: an interface 1210, a processor 1220, and a memory 1230; wherein, the interface is used for The repairing device is connected; the memory is used to store the instructions; the processor is used to execute the instructions to implement the method in the foregoing embodiment.
- the storage device may also include a bus 1260 for internal interconnection.
- the storage device is a storage die, and the repair device is a logic die; or, the storage device is a storage die, and the repair device is located on a logic die including a controller , and the repairing device is connected to the storage die through the controller.
- the embodiment of the present application further provides a repairing device, and the repairing device can be used to execute the method performed by any repairing device in the foregoing embodiments.
- the repair device may be a logic die in a memory chip.
- the repair device may be located on a logic die in a memory chip, on which a controller may also be deployed.
- the repairing device can communicate with the bus, and can also communicate with the memory die in the memory chip through the controller.
- An embodiment of the present application further provides a storage device, and the storage device can be used to execute the method executed by any of the storage devices in the foregoing embodiments.
- the memory device may be a memory die in a memory chip.
- the memory device may be located on a memory die in a memory chip.
- Embodiments of the present application further provide a logic die, where the logic die can be used to execute the method performed by any of the repair apparatuses in the foregoing embodiments.
- the logic die may include a repair device and a controller.
- the repairing device may also be located in the controller, which is not limited in this application.
- Embodiments of the present application further provide a storage die, and the memory die can be used to execute the method executed by any of the storage devices in the foregoing embodiments.
- Embodiments of the present application further provide a memory chip, where the memory chip may include a logic die and a memory die, wherein the logic die includes a repair device and a controller, and the memory die may be used to execute any one of the memory chips in the foregoing embodiments.
- the repairing device may be used to execute the method performed by any repairing device in the foregoing embodiments.
- the embodiment of the present application further provides a main chip, the main chip can communicate with the bus and the memory chip, and the main chip can be used to execute the method performed by any repairing apparatus in the foregoing embodiments.
- the main chip may include a repair device and a controller.
- the repairing device may also be located in the controller, which is not limited in this application.
- the embodiment of the present application further provides a memory chip, the memory chip can communicate with the main chip, and the memory chip can be used to execute the method executed by any of the storage devices in the foregoing embodiments.
- Embodiments of the present application further provide a controller, where the controller is located on a logic die, and the controller can be configured to forward an access instruction sent by a restoration device to a storage die, and forward an access response and restoration information sent by the storage die to the restoration device Wait.
- the controller is configured to receive an access command sent by the repair device, where the access command includes a redundant address pointing to a redundant unit in the storage die, wherein the redundant unit is used to replace the memory unit in the storage die. Failing units, the controller can forward the access instructions to the memory die.
- the controller may also receive an access response sent by the memory die, where the access response includes an access result obtained when the access request is performed based on the redundant address of the redundant unit, and the controller may include the redundant address of the redundant unit.
- the access response is forwarded to the repair device.
- the controller may be a DRAM controller.
- the above-mentioned embodiments it may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
- software it can be implemented in whole or in part in the form of a computer program product.
- the computer program product includes one or more computer instructions.
- the computer program instructions when loaded and executed on a computer, result in whole or in part of the processes or functions described herein.
- the computer may be a general purpose computer, special purpose computer, computer network, or other programmable device.
- the computer instructions may be stored in or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line) or wireless (eg, infrared, wireless, microwave, etc.).
- the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes an integration of one or more available media.
- the usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVD), or semiconductor media (eg, Solid State Disk), and the like.
Abstract
Description
Claims (20)
- 一种存储单元的访问方法,其特征在于,应用于修复装置,所述方法包括:接收包含访问地址的访问请求;所述访问地址包括存储装置中的至少一个存储单元的原始地址;根据全局修复信息,识别所述访问地址指向的存储单元中是否存在失效单元;其中,所述全局修复信息包含所述存储装置中的至少一个失效单元的原始地址与用于替代所述至少一个失效单元中每个失效单元的冗余单元的冗余地址;在识别到所述访问地址指向的存储单元中存在第一失效单元时,将所述访问地址中指向所述第一失效单元的原始地址替换为所述第一失效单元对应的第一冗余单元的冗余地址;将替换后的访问请求发送至所述存储装置。
- 根据权利要求1所述的方法,其特征在于,在所述接收包含访问地址的访问请求之前,所述方法还包括:接收所述存储装置发送的所述存储装置中存在的第二失效单元的原始地址;分配用于替代第二失效单元的第二冗余单元;将所述第二失效单元的原始地址和所述第二冗余单元的冗余地址的对应关系添加到所述全局修复信息;其中,所述第二失效单元与所述第一失效单元相同或不同。
- 根据权利要求1或2所述的方法,其特征在于,所述修复装置包括一次性可编程存储器efuse;在所述接收包含访问地址的访问请求之前,所述方法还包括:接收所述存储装置发送的所述存储装置中在封装前存在的第三失效单元的原始地址;分配用于替代所述第三失效单元的第三冗余单元;将所述第三失效单元的原始地址和所述第三冗余单元的冗余地址的对应关系存储在所述efuse中;在所述修复装置每次上电后,将从所述efuse中读取到的所述存储装置中存在的第三失效单元的原始地址和第三冗余单元的冗余地址的对应关系,添加到所述全局修复信息;其中,所述第三失效单元与所述第一失效单元相同或不同。
- 根据权利要求1-3任一所述的方法,其特征在于,所述方法还包括:接收所述存储装置发送的包含所述第一冗余单元的冗余地址的访问响应;根据全局修复信息,将所述第一冗余单元的冗余地址替换为所述第一失效单元的原始地址;将包含所述第一失效单元的原始地址的所述访问响应发送至总线。
- 根据权利要求1-4任一所述的方法,其特征在于,所述存储装置为存储裸片,所述修复装置为逻辑裸片;或者,所述存储装置为存储裸片,所述修复装置位于包含控制器的逻辑裸片上,且所述 修复装置通过所述控制器与所述存储裸片相连;或者,所述存储装置为存储芯片,所述修复装置为主芯片;或者,所述存储装置为存储芯片,所述修复装置位于包含控制器的主芯片上,且所述修复装置通过所述控制器与所述存储芯片相连。
- 根据权利要求1-5任一所述的方法,其特征在于,所述存储装置为存储裸片,所述修复装置位于包含控制器的逻辑裸片上,且所述修复装置通过所述控制器与所述存储裸片相连;所述存储裸片包括至少两个数据存储区,所述全局修复信息包括所述至少两个数据存储区中的所有失效单元的原始地址与用于替换所述至少两个数据存储区中的各个失效单元的冗余单元的冗余地址;任一存储单元为一个数据行,任一冗余单元为一个冗余行。
- 根据权利要求2或6所述的方法,其特征在于,所述第一冗余单元与所述第一失效单元位于所述存储装置中的同一扇区;或者,所述第一冗余单元与所述第一失效单元位于所述存储装置中同一块中的不同扇区;或者,所述第一冗余单元与所述第一失效单元位于所述存储装置中的不同块。
- 一种存储单元的访问方法,其特征在于,应用于存储装置,所述存储装置包括至少一个存储单元,所述方法包括:接收修复装置发送的包含访问地址的访问请求,其中,所述访问地址包括所述存储装置中的冗余单元的冗余地址;其中,所述冗余单元用于替代所述至少一个存储单元中的失效单元;访问所述冗余地址;向所述修复装置发送访问响应。
- 根据权利要求8所述的方法,其特征在于,在所述接收修复装置发送的包含访问地址的访问请求之前,所述方法还包括:在所述存储装置中的存储单元中存在失效单元时,向所述修复装置发送所述失效单元的原始地址,以使述修复装置分配用于替代所述失效单元的所述冗余单元,并将所述失效单元的原始地址和所述冗余单元的冗余地址之间的对应关系加入全局修复信息。
- 根据权利要求8或9所述的方法,其特征在于,所述存储装置为存储裸片,所述修复装置为逻辑裸片;或者,所述存储装置为存储裸片,所述修复装置位于包含控制器的逻辑裸片上,且所述修复装置通过所述控制器与所述存储裸片相连;或者,所述存储装置为存储芯片,所述修复装置为主芯片;或者,所述存储装置为存储芯片,所述修复装置位于包含控制器的主芯片上,且所述修复装置通过所述控制器与所述存储芯片相连。
- 一种存储单元的修复方法,其特征在于,应用于存储装置,所述方法包括:在所述存储装置中的存储单元中存在失效单元时,向修复装置发送所述失效单元的原始地址。
- 根据权利要求8或9所述的方法,其特征在于,所述在所述存储装置中的存储 单元中存在失效单元时,向修复装置发送所述失效单元的原始地址,包括:在所述存储装置封装前,检测所述存储装置中的存储单元中是否存在失效单元;在所述存储装置中的存储单元中存在失效单元时,向所述修复装置发送所述失效单元的原始地址。
- 一种存储单元的修复方法,其特征在于,应用于修复装置,所述方法包括:接收存储装置发送的所述存储装置中的存储单元中存在的失效单元的原始地址;分配用于替换所述失效单元的冗余单元;将所述失效单元的原始地址与所述冗余单元的冗余地址添加到全局修复信息。
- 根据权利要求13所述的方法,其特征在于,所述修复装置包括一次性可编程存储器efuse;所述接收存储装置发送的所述存储装置中的存储单元中存在的失效单元的原始址,包括:接收所述存储装置发送的所述存储装置中在封装前存在的至少一个失效单元的原始地址;所述分配用于替换所述失效单元的冗余单元,包括:分配用于替代在封装前存在的各个失效单元的冗余单元;所述将所述失效单元的原始地址与所述冗余单元的冗余地址添加到全局修复信息,包括:将所述在封装前存在的各个失效单元的原始地址和各个失效单元对应的冗余单元的冗余地址的对应关系存储在所述efuse中;在所述修复装置每次上电后,将从所述efuse中读取到的所述存储装置中存在的至少一个失效单元的原始地址和各个失效单元对应的冗余单元的冗余地址的对应关系,添加到所述全局修复信息。
- 一种修复装置,其特征在于,包括:接口,处理器和存储器;其中,所述接口用于连接存储装置;所述存储器,用于存储指令;所述处理器,用于执行指令以实现权利要求1-7,13-14任一所述的方法。
- 根据权利要求15所述修复装置,其特征在于,所述存储装置为存储裸片,所述修复装置为逻辑裸片;或者,所述存储装置为存储裸片,所述修复装置位于包含控制器的逻辑裸片上,且所述修复装置通过所述控制器与所述存储裸片相连。
- 根据权利要求15所述修复装置,其特征在于,所述存储器还包括:一次性可编程存储器efuse。
- 一种存储装置,其特征在于,包括:接口,处理器和存储器;其中,所述接口用于连接修复装置;所述存储器,用于存储指令和数据;所述处理器,用于执行指令以实现权利要求8-12任一所述的方法。
- 根据权利要求18所述存储装置,其特征在于,所述存储装置为存储裸片,所述修复装置为逻辑裸片;或者,所述存储装置为存储裸片,所述修复装置位于包含控制器的逻辑裸片上,且所述修复装置通过所述控制器与所述存储裸片相连。
- 一种存储芯片,其特征在于,包括:逻辑裸片和至少一个存储裸片;其中,所 述逻辑裸片包括修复装置和控制器,所述修复装置通过所述控制器与所述存储裸片通信;所述修复装置用于执行权利要求1-7,13-14任一所述的方法,所述存储裸片为存储装置,所述存储装置用于执行权利要求8-12任一所述的方法。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202180087793.3A CN116724355A (zh) | 2021-04-30 | 2021-04-30 | 存储单元的访问方法、修复方法、裸片和存储芯片 |
PCT/CN2021/091480 WO2022226997A1 (zh) | 2021-04-30 | 2021-04-30 | 存储单元的访问方法、修复方法、裸片和存储芯片 |
EP21938471.6A EP4307306A4 (en) | 2021-04-30 | 2021-04-30 | ACCESS METHOD FOR STORAGE UNIT, REPAIR METHOD, CHIP AND STORAGE CHIP |
US18/495,883 US20240055070A1 (en) | 2021-04-30 | 2023-10-27 | Storage unit access method, storage unit repair method, die, and memory chip |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2021/091480 WO2022226997A1 (zh) | 2021-04-30 | 2021-04-30 | 存储单元的访问方法、修复方法、裸片和存储芯片 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US18/495,883 Continuation US20240055070A1 (en) | 2021-04-30 | 2023-10-27 | Storage unit access method, storage unit repair method, die, and memory chip |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022226997A1 true WO2022226997A1 (zh) | 2022-11-03 |
Family
ID=83847479
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/091480 WO2022226997A1 (zh) | 2021-04-30 | 2021-04-30 | 存储单元的访问方法、修复方法、裸片和存储芯片 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20240055070A1 (zh) |
EP (1) | EP4307306A4 (zh) |
CN (1) | CN116724355A (zh) |
WO (1) | WO2022226997A1 (zh) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117809702A (zh) * | 2024-02-29 | 2024-04-02 | 西安紫光国芯半导体股份有限公司 | 一种三维堆叠结构及控制方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104008780A (zh) * | 2013-02-26 | 2014-08-27 | 中芯国际集成电路制造(上海)有限公司 | 存储单元的修复方法和装置 |
US20140241043A1 (en) * | 2013-02-28 | 2014-08-28 | SK Hynix Inc. | Electronic device and method for operating electronic device |
CN104205232A (zh) * | 2012-03-30 | 2014-12-10 | 英特尔公司 | 用于存储器设备的芯片上冗余修复 |
CN110010187A (zh) * | 2017-11-24 | 2019-07-12 | 三星电子株式会社 | 半导体存储器装置及其操作方法 |
US20200117558A1 (en) * | 2018-10-16 | 2020-04-16 | Micron Technology, Inc. | Intelligent post-packaging repair |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014047225A1 (en) * | 2012-09-18 | 2014-03-27 | Mosys, Inc. | Substitute redundant memory |
WO2014070160A1 (en) * | 2012-10-31 | 2014-05-08 | Hewlett-Packard Development Company, L.P. | Repairing a memory device |
US9223665B2 (en) * | 2013-03-15 | 2015-12-29 | Micron Technology, Inc. | Apparatuses and methods for memory testing and repair |
US10395750B2 (en) * | 2017-10-05 | 2019-08-27 | Dell Products, Lp | System and method for post-package repair across DRAM banks and bank groups |
-
2021
- 2021-04-30 WO PCT/CN2021/091480 patent/WO2022226997A1/zh active Application Filing
- 2021-04-30 EP EP21938471.6A patent/EP4307306A4/en active Pending
- 2021-04-30 CN CN202180087793.3A patent/CN116724355A/zh active Pending
-
2023
- 2023-10-27 US US18/495,883 patent/US20240055070A1/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104205232A (zh) * | 2012-03-30 | 2014-12-10 | 英特尔公司 | 用于存储器设备的芯片上冗余修复 |
CN104008780A (zh) * | 2013-02-26 | 2014-08-27 | 中芯国际集成电路制造(上海)有限公司 | 存储单元的修复方法和装置 |
US20140241043A1 (en) * | 2013-02-28 | 2014-08-28 | SK Hynix Inc. | Electronic device and method for operating electronic device |
CN110010187A (zh) * | 2017-11-24 | 2019-07-12 | 三星电子株式会社 | 半导体存储器装置及其操作方法 |
US20200117558A1 (en) * | 2018-10-16 | 2020-04-16 | Micron Technology, Inc. | Intelligent post-packaging repair |
Non-Patent Citations (1)
Title |
---|
See also references of EP4307306A4 * |
Also Published As
Publication number | Publication date |
---|---|
EP4307306A1 (en) | 2024-01-17 |
US20240055070A1 (en) | 2024-02-15 |
EP4307306A4 (en) | 2024-04-10 |
CN116724355A (zh) | 2023-09-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021004231A1 (zh) | 一种闪存设备中的数据存储方法及闪存设备 | |
TWI525446B (zh) | 切換式介面堆積晶粒記憶體架構 | |
CN107077303B (zh) | 分配和配置持久存储器 | |
JP6518191B2 (ja) | アドレス断片化に対するメモリ・セグメントのリマッピング | |
US8412987B2 (en) | Non-volatile memory to store memory remap information | |
CN101719103B (zh) | 基于存储设备的信息处理方法以及存储设备 | |
EP1183606B1 (en) | Address mapping in solid state storage device | |
KR102533072B1 (ko) | 블록의 상태에 따라 사용 여부를 결정하는 메모리 시스템 및 메모리 시스템의 동작 방법 | |
JP2012519900A (ja) | メモリブロックの管理 | |
KR20100116396A (ko) | 컨트롤러, 이를 포함하는 데이터 저장 장치 및 데이터 저장 시스템, 및 그 방법 | |
CN101165659A (zh) | 用于在存储装置中再分配可寻址空间的方法和设备 | |
JPH08203294A (ja) | メモリ | |
US20160246712A1 (en) | Indirection data structures implemented as reconfigurable hardware | |
CN110447075B (zh) | 多内核管芯上存储器微控制器 | |
KR20200032527A (ko) | 메모리 시스템의 동작 방법 및 메모리 시스템 | |
US20240055070A1 (en) | Storage unit access method, storage unit repair method, die, and memory chip | |
CN111679783A (zh) | 存储器控制器 | |
US20210166776A1 (en) | Memory device with a repair match mechanism and methods for operating the same | |
US11573735B2 (en) | Technologies for media management in column-addressable memory media systems | |
KR20210141156A (ko) | 페이지 장애에 기초하여 메모리 고장을 예견하고, 예견되는 메모리 고장을 관리하는 시스템의 운영 체계 핸들링 | |
US20220300165A1 (en) | Cooperative Storage Architecture | |
JP2003323352A (ja) | メモリ装置 | |
US11966610B2 (en) | Storage device capable of adjusting size of zoned namespace and operating method thereof | |
US11901032B2 (en) | Memory device and memory system capable of using redundancy memory cells | |
CN111694772A (zh) | 存储器控制器 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21938471 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180087793.3 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2021938471 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2021938471 Country of ref document: EP Effective date: 20231013 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |