Memory device and operating method thereof

By performing read-modify-write operations and pre-charge commands in the volatile memory device, the problem of row hammer phenomenon is solved, and efficient access management of volatile memory devices is achieved. , reduces protocol complexity and performance losses.

CN119068940APending Publication Date: 2024-12-03SAMSUNG ELECTRONICS CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410683819.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-10-04
Filing Date
2024-05-30
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Volatile memory devices such as DRAM are prone to row hammering when intensively accessing specific memory cell rows, resulting in data loss of adjacent memory cell rows. It is necessary to effectively manage the number of accesses of each memory cell row to prevent such phenomena.

Method used

By performing a read-modify-write (RMW) operation, the number of accesses of each storage unit row is managed with the pre-charge command, the count data is read out, and the count data is modified and updated so that there is no need to define a separate dedicated command in the memory device RMW operation between the memory controller and the device can be achieved.

Benefits of technology

It effectively prevents the occurrence of row hammer phenomenon, reduces the protocol complexity between the memory controller and the device, and can perform RMW operations in parallel when multiple precharge commands are received continuously, keeping performance not degraded.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068940A_ABST
    Figure CN119068940A_ABST
Patent Text Reader

Abstract

An example memory device includes an array of memory cells, a row hammer management circuit, and a read-modify-write (RMW) driver. The memory cell array includes a plurality of memory cell rows, and stores count data for the number of accesses to each memory cell row. The row hammer management circuit performs an RMW operation that reads count data corresponding to a target memory cell row among the memory cell rows, updates the read count data, and writes the updated count data into the memory cell array. The RMW driver generates a control signal to control RMW operation based on the pre-charge command. The target memory cell row is pre-charged after a predetermined time elapses from a point in time at which the pre-charge command is applied.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the priority of Korean Patent Application No. 10-2023-0071163 filed on June 1, 2023, and Korean Patent Application No. 10-2023-0132061 filed on October 4, 2023, in the Korean Intellectual Property Office, the disclosures of which are incorporated herein by reference in their entirety. Technical Field

[0003] One or more aspects of an embodiment of the present disclosure relate to a memory device, and more particularly, to a memory device capable of effectively processing the number of accesses to each memory cell. Background Art

[0004] Memory devices are used to store data and are classified into volatile memory devices and nonvolatile memory devices. A volatile memory device is a memory device that loses its stored data when power supply is interrupted. Among volatile memory devices, dynamic random access memory (DRAM) is used in various fields such as mobile systems, servers, graphic units, etc.

[0005] Volatile memory devices, such as DRAM, determine data based on charges stored in capacitors. However, since the charges stored in capacitors gradually leak in various ways, volatile memory devices need to periodically perform refresh operations.

[0006] Meanwhile, when a specific memory cell row is intensively accessed, a row hammer phenomenon may occur, resulting in loss of data stored in memory cells included in a memory cell row adjacent to the specific memory cell row. Therefore, it is necessary to effectively process the number of accesses to each memory cell row to prevent the occurrence of the row hammer phenomenon. Summary of the invention

[0007] The present disclosure relates to memory devices, including memory devices capable of efficiently handling the number of accesses to each memory cell.

[0008] According to the above, the memory device effectively manages the number of accesses to each memory cell row using the precharge command. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other objects and features of the present disclosure will become apparent by describing in detail embodiments of the present disclosure with reference to the attached drawings.

[0010] Figure 1 is a block diagram illustrating an example of a memory device.

[0011] Figure 2AA block diagram illustrating an example of a read-modify-write (RMW) driver.

[0012] Figure 2B A block diagram illustrating an example of a read-modify-write (RMW) driver.

[0013] Figure 3 An example diagram illustrating a precharge command.

[0014] Figure 4 is a flow chart illustrating an example of memory device operation.

[0015] Figure 5 is a block diagram illustrating an example memory system.

[0016] Figure 6 is a block diagram illustrating an example of a configuration of a memory controller.

[0017] Figure 7 is a block diagram illustrating an example of a memory device.

[0018] Figure 8 is a diagram illustrating an example of a bank array.

[0019] Fig. 9 is a diagram illustrating an example of an operation of generating an internal precharge signal.

[0020] Fig.10 is a diagram illustrating an example of arrangement of a memory device.

[0021] Fig.11A and Fig. 11B is a diagram illustrating an example of the operation of the IACU generator of the RMW driver.

[0022] Fig.12 , Fig.13A and Fig. 13B is a diagram illustrating an example of the configuration and operation of the RMW control circuit of the RMW driver.

[0023] Fig.14A is a diagram illustrating an example of a memory bank array.

[0024] Fig. 14B is a diagram illustrating an example of a memory bank array.

[0025] Fig.15 is a diagram illustrating an example of a memory bank array.

[0026] Fig.16 is a block diagram illustrating an example of a memory bank array and sense amplifiers.

[0027] Fig.17 For illustration Fig.16A diagram of an example of a portion of a first memory bank array.

[0028] Fig.18 For illustration Fig.17 A circuit diagram of an example of a bit line sense amplifier.

[0029] Fig.19 is a block diagram illustrating an example of a memory device that performs a normal read or write operation.

[0030] Fig. 20 is a block diagram illustrating an example of a memory device that performs a RMW operation.

[0031] Fig.21 is a flow chart illustrating an example of the operation of a memory device.

[0032] Fig.22A is a block diagram illustrating an example of a memory system.

[0033] Fig. 22B For illustration Fig.22A A block diagram of an example of a control logic circuit.

[0034] Fig.23A is a block diagram illustrating an example of a memory system.

[0035] Fig. 23B For illustration Fig.23A A block diagram of an example of a control logic circuit. DETAILED DESCRIPTION

[0036] The embodiments of the present disclosure will be described in detail and clearly below so that those skilled in the art can easily implement the present disclosure.

[0037] Figure 1 is a block diagram illustrating an example of a memory system 10 .

[0038] The memory system 10 can manage the number of accesses to each memory cell row to prevent a row hammer phenomenon caused by intensive accesses to a specific memory cell row. To this end, the memory device 200 can perform a read-modify-write (hereinafter referred to as "RMW") operation, which reads out count data corresponding to a target memory cell row in an activated state, modifies the read count data, and writes the modified count data again. That is, the memory device 200 can read out count data corresponding to a target memory cell row in an activated state through the RMW operation, can modify the read count data, and can write the modified count data again to update the count data, so that the number of accesses to each memory cell row can be managed.

[0039] In this case, the memory device 200 may perform a RMW operation based on a precharge command received from the memory controller 100. The precharge command according to the present embodiment may be issued a predetermined amount of time earlier than a normal precharge command. The memory device 200 may perform a RMW operation based on the precharge command, and then may perform a precharge operation. As described above, the memory device 200 according to the present embodiment may effectively perform a RMW operation without defining a separate dedicated command for the RMW operation between the memory controller 100 and the memory device 200.

[0040] refer to Figure 1 , the memory system 10 may include a memory controller 100 and a memory device 200 .

[0041] The memory controller 100 may control the memory device 200. As an example, the memory controller 100 may control the memory device 200 in response to a request of a processor supporting various applications, such as a server application, a personal computer (PC) application, a mobile application, etc. For example, the memory controller 100 may be included in a host including a processor, and may control the memory device 200 in response to a request of the processor.

[0042] The memory controller 100 may transmit a command and / or an address to the memory device 200 to control the memory device 200. In addition, the memory controller 100 may transmit data to the memory device 200 or may receive data from the memory device 200.

[0043] The memory device 200 may receive data from the memory controller 100 and store the data. The memory device 200 may read out the stored data in response to a request of the memory controller 100, and may transmit the read-out data to the memory controller 100.

[0044] In some embodiments, the memory device 200 may be a memory device including a volatile memory cell. For example, the memory device 200 may be various DRAM devices, such as double data rate synchronous dynamic random access memory (DDR SDRAM), DDR2 SDRAM, DDR3 SDRAM, DDR4 SDRAM, DDR5 SDRAM, low power double data rate (LPDDR) SDRAM, LPDDR2 SDRAM, LPDDR3 SDRAM, LPDDR4 SDRAM, LPDDR4X SDRAM, LPDDR5 SDRAM, graphic double data rate synchronous graphic random access memory (GDDR SGRAM), GDDR2 SGRAM, GDDR3 SGRAM, GDDR4 SGRAM, GDDR5 SGRAM, GDDR6 SGRAM, etc.

[0045] In some embodiments, the memory device 200 may be a memory device stacked with DRAM dies, such as high bandwidth memory (HBM), HBM2, HBM3, etc.

[0046] In some embodiments, the memory device 200 may be a memory module, such as a dual in-line memory module (DIMM). For example, the memory device 200 may be a registered DIMM (RDIMM), a load reduced DIMM (LRDIMM), an unbuffered DIMM (UDIMM), a fully buffered DIMM (FB-DIMM), or a small outline DIMM (SO-DIMM), however, this is only an example. In some embodiments, the memory device 200 may be any other memory module (e.g., a single in-line memory module (SIMM)).

[0047] In some embodiments, the memory device 200 may be an SRAM device, a NAND flash device, a NOR flash device, an RRAM device, a FRAM device, a TRAM device, an MRAM device, or the like.

[0048] The memory device 200 may include a memory cell array 310 , a row hammer (RH) management circuit 500 , and a RMW driver 600 .

[0049] The memory cell array 310 may include a plurality of memory banks Bank 1 to Bank n, each of which may include memory cells storing data. In the present disclosure, for ease of explanation, it is assumed that each memory bank includes a DRAM cell, however, this is merely an example. In some implementations, each of the memory banks Bank 1 to Bank n may be implemented to include a volatile memory cell in addition to the DRAM cell. Furthermore, the memory banks Bank 1 to Bank n may be implemented to include the same type of memory cells, or may be implemented to include different types of memory cells.

[0050] The memory cell array 310 may include a plurality of memory cell rows. Here, the memory cell row may refer to memory cells included in one row.

[0051] In some embodiments, a memory cell row may be defined for each memory bank. As an example, a memory bank may be connected to multiple word lines, and memory cells connected to the same word line may constitute the same memory cell row. In this case, each memory bank may include multiple memory cell rows.

[0052] In some embodiments, a memory cell row may be defined for each sub-array block. As an example, a memory bank may include multiple sub-array blocks, and memory cells connected to the same word line among the memory cells of each sub-array block may constitute the same memory cell row. In this case, each sub-array block may include multiple memory cell rows.

[0053] In some embodiments, a memory cell row may be defined as memory cells including memory banks that are different from each other. As an example, two or more memory banks may share the same word line, and memory cells connected to the same word line and included in memory banks that are different from each other may constitute the same memory cell row, however, this is merely an example. In some embodiments, a memory cell row may be defined in various ways.

[0054] Some memory cells of the memory cell row may be used as counting cells. Here, the counting cell may refer to a memory cell that stores the number of accesses to the corresponding memory cell row as counting data. Each memory cell row may include a counting cell, however, this is only an example. In some embodiments, only some memory cell rows may include a counting cell. In some embodiments, counting cells corresponding to the memory cell rows may be provided in separate memory banks.

[0055] The row hammer management circuit 500 may manage the number of accesses to each memory cell row based on a command applied from the memory controller 100. As an example, when an activation command is issued from the memory controller 100 and a word line is activated, the row hammer management circuit 500 may count the number of accesses to the target memory cell row corresponding to the activated word line. Then, the row hammer management circuit 500 may store count data as the number of accesses in a counting unit corresponding to the target memory cell row.

[0056] In some embodiments, the row hammer management circuit 500 may manage the number of accesses to each memory cell row through a RMW operation. As an example, when an activation command for a target memory cell row is issued from the memory controller 100, the row hammer management circuit 500 may perform a RMW operation that reads count data from a count cell of the target memory cell row, adds "1" to the read count data to generate updated count data, and writes the updated count data to the count cell of the target memory cell row again.

[0057] In some embodiments, the row hammer management circuit 500 may perform a RMW operation based on a precharge command applied from the memory controller 100. That is, the precharge command may be used not only as a command to precharge a word line of a target memory cell row, but also as a command to update the number of accesses to the target memory cell row. Therefore, there is no need to define a separate dedicated command to perform a RMW operation between the memory controller 100 and the memory device 200, and thus the complexity of the protocol between the memory controller 100 and the memory device 200 may be reduced.

[0058] In some embodiments, the memory device 200 may perform a RMW operation and a precharge operation based on a precharge command, and the RMW operation may be performed before the precharge operation.

[0059] For example, when a precharge command is issued from the memory controller 100, the row hammer management circuit 500 may perform an RMW operation on the target memory cell row within a predetermined time based on the precharge command. Here, the predetermined time for performing the RMW operation may be referred to as an active count update time. After the active count update time has passed, the memory device 200 may perform a precharge operation on the target memory cell row.

[0060] The RMW driver 600 may generate a signal required to perform an RMW operation in response to a precharge command applied thereto by the memory controller 100. Specifically, when precharge commands for different memory banks are continuously received from the memory controller 100, the RMW driver 600 may generate a signal required to perform an RMW operation on each memory bank. Therefore, each memory bank may perform an RMW operation in parallel, and as a result, an RMW operation may be effectively performed on memory banks different from each other.

[0061] In some embodiments, the RMW driver 600 may generate an internal activation count update signal based on the precharge command. In particular, when the precharge commands for different memory banks are received successively, the RMW driver 600 may generate an internal activation count update signal for each memory bank.

[0062] In some embodiments, the RMW driver 600 may generate core signals based on the internal activation count update signal to control the RMW operation. For example, the core signals may include an input / output (I / O) sense amplifier enable signal, a column select line enable signal, and the like.

[0063] Specifically, when precharge commands for different memory banks are received continuously, the RMW driver 600 can generate a core control signal for each memory bank based on the internal activation count update signal of each memory bank. Therefore, when precharge commands for different memory banks are received continuously, the RMW operation of each memory bank can be performed in parallel. Therefore, even if precharge commands for different memory banks are received continuously at short time intervals, the RMW operation of each memory bank can be effectively performed without performance degradation.

[0064] As described above, the memory device 200 can perform the RMW operation based on the precharge command without defining a separate dedicated command for the RMW operation between the memory controller 100 and the memory device 200. In addition, even when the precharge commands for different memory banks are received continuously, the memory device 200 can perform the RMW operation on each memory bank in parallel. Therefore, the RMW operation can be performed without reducing the performance and increasing the complexity of the protocol between the memory controller 100 and the memory device 200.

[0065] Figure 2A is a block diagram of an example of an RMW driver 600A. Figure 2A The RMW driver 600A can be Figure 1 RMW drives 600.

[0066] refer to Figure 2A The RMW driver 600A may include an internal activation count update signal generator (hereinafter referred to as an “IACU generator”) 610 and a plurality of RMW control circuits 620_1 to 620_n.

[0067] The IACU generator 610 may receive the precharge command PRE, and may generate an internal activation count update signal (hereinafter referred to as an 'IACU signal') based on the received precharge command PRE.

[0068] The IACU generator 610 may continuously receive first to nth precharge commands PRE1 to PREn corresponding to different memory banks. In this case, the IACU generator 610 may generate first to nth internal activation count update signals IACU1 to IACUn corresponding to the first to nth precharge commands PRE1 to PREn in parallel. The first to nth internal activation count update signals IACU1 to IACUn may correspond to different memory banks, respectively.

[0069] Each of the first through nth RMW control circuits 620_1 through 620_n may be implemented to generate a core signal for one memory bank.

[0070] As an example, the first RMW control circuit 620_1 may receive the first internal activation count update signal IACU1 corresponding to the first memory bank and may generate core signals RD_CTRL1_1, RD_CTRL1_2, WR_CTRL1_1, and WR_CTRL1_2 to control the RMW operation of the first memory bank.

[0071] Here, the first_first read control signal RD_CTRL1_1 and the first_second read control signal RD_CTRL1_2 may be control signals for reading out count data stored in a count unit of a target memory cell row included in the first memory bank. The first_first write control signal WR_CTRL1_1 and the first_second write control signal WR_CTRL1_2 may be control signals for writing updated count data into the count unit again. The core signals RD_CTRL1_1, RD_CTRL1_2, WR_CTRL1_1, and WR_CTRL1_2 for controlling the RMW operation of the first memory bank may be referred to as a first RMW core signal group RMW_CSG1.

[0072] Similarly, the nth RMW control circuit 620_n may receive the nth internal activation count update signal IACUn corresponding to the nth memory bank and may generate an nth RMW core signal group RMW_CSGn.

[0073] As described above, the RMW driver 600A according to the present embodiment can generate the RMW core signal group RMW_CSG to perform the RMW operation on each memory bank. Specifically, even when the precharge command PRE corresponding to different memory banks is received continuously, the RMW driver 600A can generate the RMW core signal group RMW CSG for each memory bank in parallel. Therefore, the RMW operation of each memory bank can be effectively performed without reducing the performance.

[0074] Figure 2B is a block diagram of an example of an RMW driver 600B. Figure 2B The RMW driver 600B can be Figure 1 RMW drives 600. Figure 2B The RMW drive 600B is similar to Figure 2A The RMW drive is 600A. Figure 2B In the drawings, the same or similar reference numerals denote Figure 2A The same or similar elements are described herein, and thus detailed descriptions of the same or similar elements will be omitted.

[0075] One RMW control circuit corresponds to a memory bank Figure 2A The structure of the RMW driver 600A is different. Figure 2BThe RMW driver 600B may have a structure in which one RMW control circuit corresponds to a plurality of memory banks.

[0076] refer to Figure 2B , each of the RMW control circuits 620_1 to 620_k may correspond to one memory bank group BG, and the one memory bank group BG may include m memory banks ("m" is a natural number equal to or greater than 2).

[0077] As an example, the first RMW control circuit 620_1 may correspond to the first memory bank group BG1, and the first memory bank group BG1 may include m memory banks. In this case, the first RMW control circuit 620_1 may receive internal activation count update signals IACU1 to IACUm for the m memory banks in parallel, and may generate m RMW core signal groups RMW_CSG1 to RMW_CSGm corresponding to the m memory banks, respectively.

[0078] Similarly, the kth RMW control circuit 620_k may correspond to the kth memory bank group BGk. The kth RMW control circuit 620_k may receive internal activation count update signals IACU1 to IACUm for m memory banks in parallel and may generate m RMW core signal groups RMW_CSG1 to RMW_CSGm respectively corresponding to the m memory banks.

[0079] Figure 3 2 is a diagram illustrating an example of a precharge command. For ease of explanation, it is assumed that a common memory device and a memory device 200 according to the present disclosure (refer to Figure 1 ) receives the activation command ACT at the same first time point t1. In addition, it is assumed that the normal memory device receives the activation count update command ACU at the second time point t2.

[0080] refer to Figure 3 , the normal memory device receives an activation command ACT at a first time point t1. Then, the normal memory device receives an activation count update command ACU at a second time point t2, and performs an RMW operation to update the activation count. Then, the normal memory device receives a precharge command Normal PRE at a third time point t3, and therefore, the normal memory device performs a precharge operation at the third time point t3 in response to the received precharge command Normal PRE.

[0081] On the other hand, the memory device 200 according to the present disclosure may receive the activation command ACT at the first time point t1, and may receive the precharge command PRE at the second time point t2 earlier than the third time point t3. That is, the memory device 200 according to the present disclosure may receive the precharge command PRE at a time point earlier than the ordinary memory device by the activation count update time tACU. In the present embodiment, the activation count update time tACU may be a predetermined time, and may be the time required to perform the RMW operation on each memory bank.

[0082] In this case, the memory device 200 according to the present disclosure may perform an RMW operation on each memory bank based on the precharge command PRE received at the second time point t2. Then, the memory device 200 may perform a precharge operation at a third time point t3 after the activation count update time tACU has passed since the second time point t2. That is, the precharge operation may be performed by delaying the activation count update time tACU from the time point of receiving the precharge command PRE.

[0083] As described above, according to a predetermined protocol, the memory device 200 according to the present disclosure may receive a precharge command PRE at a time point earlier than the activation count update time tACU of a normal memory device. In addition, the memory device 200 according to the present disclosure may not only perform a precharge operation based on the precharge command PRE, but may also perform an RMW operation. In this case, there is no need to receive a separate dedicated command required to perform an RMW operation from the host, and therefore, the complexity of the protocol between the memory controller 100 and the memory device 200 may be reduced.

[0084] Figure 4 is a flow chart illustrating an example of the operation of a memory device. Figure 4 The described memory device may correspond to Figure 1 and Figure 3 A memory device 200 is provided.

[0085] In operation S10, the memory device 200 may receive an active command ACT. Thus, a word line corresponding to a target memory cell row may be activated. Then, according to an embodiment, a read operation or a write operation may be performed on the target memory cell row.

[0086] In operation S20, the memory device 200 may receive a precharge command PRE. The memory device 200 may receive the precharge command PRE at a time point earlier than a general memory device by an activation count update time tACU.

[0087] In operation S30, the memory device 200 may generate an internal activation count update signal IACU based on the precharge command PRE. When precharge commands PRE corresponding to different banks are continuously received, the memory device 200 may generate internal activation count update signals IACU corresponding to the continuously received precharge commands PRE in parallel.

[0088] In operation S40, the memory device 200 may perform an RMW operation based on the internal activation count update signal IACU. When the precharge commands PRE corresponding to different memory banks are continuously received, the memory device 200 may perform RMW operations on different memory banks in parallel based on the internal activation count update signal IACU generated in parallel. That is, the RMW operations on different memory banks may be performed substantially simultaneously.

[0089] In operation S50, the memory device 200 may perform a precharge operation. In this case, the precharge operation may be performed at a time point after an activation count update time tACU has elapsed since the precharge command PRE was received. Thus, the word line corresponding to the target memory cell row may be deactivated.

[0090] As described above, the memory device 200 according to the present disclosure can perform the RMW operation based on the precharge command. In addition, even when the precharge commands for different memory banks are received continuously, the memory device 200 according to the present disclosure can perform the RMW operation of each memory bank in parallel. Therefore, the RMW operation can be effectively performed without reducing the performance and increasing the complexity of the protocol between the memory controller 100 and the memory device 200.

[0091] Figure 5 is a block diagram illustrating an example of the memory system 20 . Figure 5 The memory system 20 is similar to Figure 1 The memory system 10. Figure 5 In the drawings, the same or similar reference numerals denote Figure 1 The present invention relates to the same or similar elements in the present invention, and thus a detailed description of the same elements will be omitted.

[0092] refer to Figure 5 , the memory system 20 may include a memory controller 100 and a memory device 200 , and the memory device 200 may include a memory cell array 310 , a row hammer management circuit 500 , a RMW driver 600 , and a control logic circuit 210 .

[0093] The memory controller 100 may transmit a clock signal CK, a command CMD, and an address ADDR to the memory device 200. When a data signal DQ is read out from the memory device 200, the memory controller 100 may receive a data strobe signal DQS from the memory device 200.

[0094] The control logic circuit 210 of the memory device 200 may control the operation of the memory device 200. As an example, the control logic circuit 210 may generate control signals to allow the memory device 200 to perform a write operation, a read operation, a normal refresh operation, and a hammer refresh operation.

[0095] Here, the hammer refresh operation may mean an operation of reducing a row hammer phenomenon of refreshing adjacent memory cell rows adjacent to an intensively accessed aggressor memory cell row. The normal refresh operation may mean an operation of sequentially refreshing memory cell rows of a memory cell array.

[0096] like Figure 5 As shown, the control logic circuit 210 and the RMW driver 600 may be independently implemented as different circuits.

[0097] Figure 6 is a block diagram illustrating an example of the configuration of the memory controller 100 . Figure 6 The memory controller 100 may correspond to, for example Figure 1 or Figure 5 A memory controller 100 is provided.

[0098] refer to Figure 6 , the memory controller 100 may include a CPU 120, an RFM control logic 110, a refresh logic 130, a host interface 140, a scheduler 150, and a memory interface 160, which are connected to each other through a bus.

[0099] The CPU 120 may control the overall operation of the memory controller 100. As an example, the CPU 120 may control each of the RFM control logic 110, the refresh logic 130, the host interface 140, the scheduler 150, and the memory interface 160.

[0100] The RFM control logic 110 may generate a refresh management (RFM) command for a memory cell row associated with a row hammer among memory cell rows of the memory device 200. As an example, when it is determined that a specific memory cell row is intensively accessed, the RFM control logic 110 may generate a command for refreshing adjacent memory cell rows adjacent to the attacker memory cell row.

[0101] The refresh logic 130 may generate a command to perform a normal refresh operation on the memory cell rows of the memory device 200. As an example, the refresh logic 130 may generate an auto-refresh command to sequentially refresh the memory cell rows according to a refresh cycle.

[0102] The host interface 140 may perform an interface connection with a host. The memory interface 160 may perform an interface connection with the memory device 200.

[0103] The scheduler 150 may manage the scheduling and transmission of the command sequence generated by the memory controller 100. Specifically, the scheduler 150 according to the present embodiment may transmit the precharge command PRE earlier than the transmission timing of the normal precharge command by the activation count update time tACU. Therefore, the precharge command PRE may be used as a command to perform not only the precharge operation but also the RMW operation.

[0104] Figure 7 is a block diagram illustrating an example of a memory device 200 . Figure 7 The memory device 200 may correspond to Figure 1 and Figure 5 A memory device 200 is provided.

[0105] refer to Figure 7 , the memory device 200 may include a control logic circuit 210, an address register 220, a memory bank control logic 230, a refresh control circuit 400, a row address multiplexer 240, a column address latch 250, a row decoder 260, a column decoder 270, a memory cell array 310, a sense amplifier portion 285, an input / output (I / O) gating circuit 290, an ECC engine 350, a data I / O buffer 320, a row hammer management circuit 500, and a RMW driver 600.

[0106] The memory cell array 310 may include a plurality of bank arrays 310_1 to 310_n. Each of the bank arrays 310_1 to 310_n may include a plurality of memory cells. As an example, each memory cell may be formed at a point where a corresponding word line intersects a corresponding bit line.

[0107] The row decoder 260 may include a plurality of sub-row decoders 260_1 to 260_n. Each of the sub-row decoders 260_1 to 260_n may be connected to a corresponding bank array among the bank arrays 310_1 to 310_n.

[0108] The sense amplifier part 285 may include a plurality of sense amplifiers 285_1 to 285 — n. Each of the sense amplifiers 285_1 to 285 — n may be connected to a corresponding bank array among the bank arrays 310_1 to 310 — n.

[0109] The column decoder 270 may include a plurality of sub-column decoders 270_1 to 270_n. Each of the sub-column decoders 270_1 to 270_n may be connected to a corresponding bank array among the bank arrays 310_1 to 310_n via a corresponding sense amplifier among the sense amplifiers 285_1 to 285_n.

[0110] The bank arrays 310_1 to 310_n, the sense amplifiers 285_1 to 285_n, the column decoders 270_1 to 270_n, and the row decoders 260_1 to 260_n may form a plurality of banks. As an example, the first bank array 310_1, the first sense amplifier 285_1, the first column decoder 270_1, and the first row decoder 260_1 may form a first bank.

[0111] The address register 220 can be obtained from the memory controller 100 (refer to Figure 1 ) receives an address ADDR including a bank address BANK_ADDR, a row address ROW_ADDR, and a column address COL_ADDR. The address register 220 may provide the received bank address BANK_ADDR to the bank control logic 230, may provide the received row address ROW_ADDR to the row address multiplexer 240, and may provide the received column address COL_ADDR to the column address latch 250. In addition, the address register 220 may provide the bank address BANK_ADDR and the row address ROW_ADDR to the row hammer management circuit 500 and the RMW driver 600.

[0112] The bank control logic 230 may generate a bank control signal in response to the bank address BANK_ADDR. As an example, a row decoder corresponding to the bank address BANK_ADDR among the row decoders 260_1 to 260_n may be activated in response to the bank control signal. A column decoder corresponding to the bank address BANK_ADDR among the column decoders 270_1 to 270_n may be activated in response to the bank control signal.

[0113] The row address multiplexer 240 may receive the row address ROW_ADDR from the address register 220 and may receive the refresh row address REF_ADDR from the refresh control circuit 400. The row address multiplexer 240 may selectively output the row address ROW_ADDR or the refresh row address REF_ADDR as the row address RA. The row address RA output from the row address multiplexer 240 may be applied to each of the row decoders 260_1 to 260_n.

[0114] The refresh control circuit 400 may sequentially increase or decrease the refresh row address REF_ADDR in the normal refresh mode in response to the refresh signal from the control logic circuit 210 .

[0115] The refresh control circuit 400 may receive the hammer address HADDR in the hammer refresh mode. The refresh control circuit 400 may output an address of a neighboring memory cell row neighboring the attacker memory cell row as the refresh row address REF_ADDR based on the hammer address HADDR.

[0116] A row decoder selected among the row decoders 260_1 to 260_n by the bank control logic 230 may activate a word line corresponding to the row address RA output from the row address multiplexer 240. As an example, the selected row decoder may apply a word line driving voltage to a word line corresponding to the row address.

[0117] The column address latch 250 may receive the column address COL_ADDR from the address register 220 and may temporarily store the received column address COL_ADDR. In addition, for example, in a burst mode, the column address latch 250 may gradually increase the received column address COL_ADDR. The column address latch 250 may apply the column address COL_ADDR′ obtained by temporarily storing or gradually increasing the column address COL_ADDR to each of the column decoders 270_1 to 270_n.

[0118] The column decoders activated by the bank control logic 230 in the column decoders 270_1 to 270 — n may activate sense amplifiers corresponding to the bank address BANK_ADDR and the column address COL_ADDR through the I / O gating circuit 290 .

[0119] The I / O gating circuit 290 may include a circuit for performing gating input and outputting data. In addition, the I / O gating circuit 290 may include a data latch for storing a codeword output from the memory bank arrays 310_1 to 310_n and a write driver for writing data into the memory bank arrays 310_1 to 310_n.

[0120] In some embodiments, the codeword CW read out from a selected memory bank array among the memory bank arrays 310_1 to 310_n during a read operation may be sensed by a sense amplifier corresponding to the selected memory bank array and may be stored in a data latch of the I / O gating circuit 290. In addition, the codeword CW stored in the data latch may be ECC decoded by the ECC engine 350 and may be provided as data DTA to the data I / O buffer 320. The data I / O buffer 320 may generate a data signal DQ based on the data DTA and may provide the data signal DQ to the memory controller 100 together with the data strobe signal DQS.

[0121] In some embodiments, during a write operation, data DTA to be written into a selected memory bank array among the memory bank arrays 310_1 to 310_n may be received by the data I / O buffer 320 as a data signal DQ. The data I / O buffer 320 may convert the data signal DQ into data DTA and may provide the data DTA to the ECC engine 350. The ECC engine 350 may generate a parity bit or parity data based on the data DTA and may provide a codeword CW including the data DTA and the parity bit to the I / O gating circuit 290. The I / O gating circuit 290 may write the codeword CW into the selected memory bank array.

[0122] The data I / O buffer 320 may convert the data signal DQ into data DTA during a write operation and may provide the data DTA to the ECC engine 350. The data I / O buffer 320 may convert the data DTA provided from the ECC engine 350 into the data signal DQ during a read operation.

[0123] The ECC engine 350 may perform an ECC encoding operation on the data DTA during a write operation. The ECC engine 350 may perform an ECC decoding operation on the codeword CW during a read operation.

[0124] In addition, the ECC engine 350 may perform ECC encoding and ECC decoding operations on the count data CNTD provided from the row hammer management circuit 500 .

[0125] The control logic circuit 210 may control the operation of the memory device 200. As an example, the control logic circuit 210 may generate a control signal to allow the memory device 200 to perform a write operation, a read operation, a normal refresh operation, and a hammer refresh operation. The control logic circuit 210 may include a command decoder 211 that decodes a command CMD provided from the memory controller 100 and a mode register set (MRS) 212 that sets an operation mode of the memory device 200.

[0126] The command decoder 211 may decode the command CMD to generate internal command signals, such as an internal activation signal IACT, an internal precharge signal IPRE, an internal read signal IRD, an internal write signal IWR, etc. In addition, the command decoder 211 may decode a chip select signal and a command / address signal to generate a control signal corresponding to the command CMD.

[0127] In some embodiments, when generating the internal precharge signal IPRE, the command decoder 211 may delay the precharge command PRE received from the memory controller 100 by a predetermined time. That is, the command decoder 211 may generate the internal precharge signal IPRE after a predetermined time has passed from the time point when the command decoder 211 receives the precharge command PRE from the memory controller 100. Here, the predetermined time may correspond to the activation count update time tACU. Therefore, the word line activated by the activation command ACT may remain in an activated state during the RMW operation.

[0128] The RMW driver 600 may receive a precharge command PRE from the memory controller 100. The RMW driver 600 may generate an RMW core signal group RMW_CSG to perform an RMW operation on each memory bank. The RMW driver 600 may include an IACU generator 610 and RMW control circuits 620_1 to 620_n.

[0129] The IACU generator 610 may continuously receive first to nth precharge commands PRE1 to PREn corresponding to different memory banks. In this case, the IACU generator 610 may generate first to nth internal activation count update signals IACU1 to IACUn corresponding to the first to nth precharge commands PRE1 to PREn in parallel.

[0130] As reference Figure 2A As described, each of the first to kth RMW control circuits 620_1 to 620_k may be implemented to generate a core signal for one bank. As an example, the first RMW control circuit 620_1 may receive a first internal activation count update signal IACU1 corresponding to a first bank and may be referred to as a first RMW core signal group RMW_CSG1 to control the RMW operation of the first bank.

[0131] As reference Figure 2B As described, each of the first RMW control circuit 620_1 to the kth RMW control circuit 620_k may correspond to a memory bank group BG, and a memory bank group BG may include m memory banks. As an example, the first RMW control circuit 620_1 may receive internal activation count update signals IACU1 to IACUm for m memory banks in parallel, and may generate m RMW core signal groups RMW_CSG1 to RMW_CSGm corresponding to the m memory banks in parallel.

[0132] Figure 8 is a view showing an example of the memory bank array 310_1. Figure 8 The bank array 310_1 may correspond to, for example Figure 7 The first memory bank array 310_1.

[0133] refer to Figure 8 , the first memory bank array 310_1 may include a plurality of word lines WL0 to WLm, a plurality of bit lines BL0 to BLn, and a plurality of memory cells MC located at points where the word lines WL0 to WLm intersect the bit lines BL0 to BLn.

[0134] Each memory cell MC may be a DRAM cell. As an example, each memory cell MC may include a cell transistor connected to a word line and a bit line and a cell capacitor connected to the cell transistor.

[0135] The word lines extending in the row direction may be referred to as rows of the first memory bank array 310_1. The bit lines extending in the column direction may be referred to as columns of the first memory bank array 310_1.

[0136] In the present disclosure, memory cells connected to the same word line may be referred to as memory cell rows. In the case of intensive access to a particular memory cell row, the intensively accessed memory cell row may be referred to as an aggressor memory cell row. A memory cell row adjacent to an aggressor memory cell row may be referred to as an adjacent memory cell row. As an example, when access is concentrated on a first word line, a first memory cell row corresponding to the first word line may be referred to as an aggressor memory cell row, and a zeroth memory cell row and a second memory cell row adjacent to the first memory cell row may be referred to as adjacent memory cell rows.

[0137] Fig. 9 is a diagram illustrating an example of an operation of generating an internal precharge signal IPRE.

[0138] refer to Fig. 9 , memory device 200 (reference Figure 5 ) can be obtained from the memory controller 100 (reference Figure 5 ) receives an activation command ACT. In this case, the control logic circuit 210 (refer to Figure 7 ) may generate an internal activation signal IACT based on the active command ACT at a first time point t1. A selected word line may be activated in response to the internal activation signal IACT.

[0139] Then, the memory device 200 may receive a precharge command PRE from the memory controller 100. The control logic circuit 210 may generate an internal precharge signal IPRE based on the received precharge command PRE. In this case, the control logic circuit 210 may generate an internal precharge signal IPRE at a third time point t3 after the activation count update time tACU from the second time point t2 at which the precharge command PRE is received. That is, the control logic circuit 210 may delay the precharge operation by the activation count update time tACU. Therefore, the activation count update time tACU required to perform the RMW operation on the memory cell row corresponding to the selected word line may be ensured.

[0140] After performing the RMW operation on the memory cell row corresponding to the selected word line, the control logic circuit 210 may generate the internal precharge signal IPRE at the third time point t3, and thus, the selected word line may be precharged.

[0141] Fig.10 is a diagram showing an example of the arrangement of the memory device 200 . Fig.10 The memory device 200 may correspond to, for example Figure 1 , Figure 5 and Figure 7 For convenience of explanation, it is assumed that the memory device 200 includes thirty-two memory banks Bank 1 to Bank 32 and four memory banks form one memory bank group BG. In addition, it is assumed that one RMW control circuit corresponds to one memory bank group.

[0142] Reference Fig.10 , the RMW driver 600 may include an IACU generator 610 and first to eighth RMW control circuits 620_1 to 620_8.

[0143] The first memory bank Bank 1 to the fourth memory bank Bank 4 may form a first memory bank group BG1. As an example, the first memory bank Bank 1 and the third memory bank Bank 3 may be disposed at an upper portion of a substrate, and the second memory bank Bank 2 and the fourth memory bank Bank 4 may be disposed at a lower portion of the substrate. The first RMW control circuit 620_1 may correspond to the first memory bank group BG1, and may be disposed adjacent to the first memory bank group BG1. As an example, the first RMW control circuit 620_1 may be disposed at the center between the first memory bank Bank 1 and the third memory bank Bank 3 and the second memory bank Bank 2 and the fourth memory bank Bank 4.

[0144] Similarly, the fifth memory bank Bank 5 to the eighth memory bank Bank 8 can form a second memory bank group BG2, and the second RMW control circuit 620_2 can be set adjacent to the second memory bank group BG2. Similarly, the twenty-ninth memory bank Bank 29 to the thirty-second memory bank Bank 32 can form an eighth memory bank group BG8, and the eighth RMW control circuit 620_8 can be set adjacent to the eighth memory bank group BG8.

[0145] The region where the IACU generator 610 and the first to eighth RMW control circuits 620_1 to 620_8 are disposed may correspond to a peripheral region PERI of the substrate.

[0146] The IACU generator 610 may generate an internal activation count update signal IACU for each memory bank. As an example, the IACU generator 610 may receive a first precharge command PRE1 for a first memory bank Bank 1, and may generate a first internal activation count update signal IACU1 corresponding to the first memory bank Bank 1. When the precharge command for the memory bank is continuously received, the IACU generator 610 may generate an internal activation count update signal IACU corresponding to each memory bank.

[0147] Each of the first to eighth RMW control circuits 620_1 to 620_8 may receive at least one internal activation count update signal IACU for a corresponding memory bank group from the IACU generator 610. Each of the first to eighth memory bank groups BG1 to BG8 may generate an RMW core signal group RMW_CSG for each memory bank based on the received internal activation count update signal IACU.

[0148] As an example, the first RMW control circuit 620_1 may receive at least one of the first to fourth internal activation count update signals IACU1 to IACU4. The first RMW control circuit 620_1 may generate an RMW control signal group RMW_CSG for a corresponding memory bank based on the received at least one internal activation count update signal.

[0149] Fig.11A and Fig. 11B 1 is a diagram showing an example of the operation of the IACU generator of the RMW driver. In detail, Fig.11A The operation of the IACU generator generating an internal activation count update signal IACU based on a precharge command PRE is shown. Fig. 11B The operation of the IACU generator generating a plurality of internal activation count update signals IACU in parallel based on a plurality of precharge commands PRE received in succession is shown. Fig.11Aand Fig. 11B In the description, it is assumed that the IACU generator and the memory device are as follows Fig.10 Arranged as in.

[0150] refer to Fig.10 and Fig.11A , the memory device 200 may receive a first precharge command PRE1 for the first memory bank Bank1 at a first time point t1. In this case, the IACU generator 610 may transition a level of the first internal activation count update signal IACU1 from a low level to a high level based on the first precharge command PRE1.

[0151] Then, when the activation count update time tACU elapses, the control logic circuit 210 may generate the first internal precharge signal IPRE1 at the second time point t2. In this case, the IACU generator 610 may change the level of the first internal activation count update signal IACU1 from a high level to a low level at the second time point t2 based on the first internal precharge signal IPRE1.

[0152] Therefore, the first internal activation count update signal IACU1 may be maintained in a high state during the activation count update time tACU.

[0153] refer to Fig.10 and Fig. 11B , the memory device 200 may continuously receive first to fourth precharge commands PRE1 to PRE4 for first to fourth memory banks Bank 1 to Bank 4 of the first memory bank group BG1. In this case, according to some implementations, the time interval between the precharge commands may be very short.

[0154] The IACU generator 610 may transition levels of the first to fourth internal activation count update signals IACU1 to IACU4 from a low level to a high level based on the successively received first to fourth precharge commands PRE1 to PRE4 .

[0155] As an example, the IACU generator 610 may change the levels of the first to fourth internal activation count update signals IACU1 to IACU4 from a low level to a high level at the first to fourth time points t1 to t4, respectively. In this case, each of the first to fourth internal activation count update signals IACU1 to IACU4 may maintain a high level during the activation count update time tACU.

[0156] Therefore, in the case of continuously receiving the first precharge command PRE1 to the fourth precharge command PRE4 for the first memory bank Bank 1 to the fourth memory bank Bank 4, the IACU generator 610 can generate the first internal activation count update signal IACU1 to the fourth internal activation count update signal IACU4 in parallel for the first memory bank Bank 1 to the fourth memory bank Bank 4. In this case, the sections in which the first internal activation count update signal IACU1 to the fourth internal activation count update signal IACU4 respectively maintain a high level can overlap with each other.

[0157] Fig.12 , Fig.13A and Fig. 13B 1 is a diagram showing an example of the configuration and operation of the RMW control circuit of the RMW driver. In detail, Fig.12 2 shows a configuration of a first RMW control circuit 620_1 according to an embodiment of the present disclosure. Fig.13A The operation of the first RMW control circuit 620_1 generating an RMW core signal group RMW_CSG based on an internal activation count update signal IACU is shown. Fig. 13B An operation of the first RMW control circuit 620_1 generating the first to fourth RMW core signal groups RMW_CSG1 to RMW_CSG4 based on the first to fourth internal activation count update signals IACU1 to IACU4 is shown.

[0158] For the sake of illustration, assume that Fig.12 , Fig.13A and Fig. 13B The RMW control circuit and memory devices such as Fig.10 In addition, it is assumed that the internal activation count update signal IACU is generated as in FIG. 11 .

[0159] refer to Fig.12 The first memory bank group BG1 may include a first memory bank array 310_1, a second memory bank array 310_2, a third memory bank array 310_3 and a fourth memory bank array 310_4, a first row decoder 260_1, a second row decoder 260_2, a third row decoder 260_3 and a fourth row decoder 260_4, a first column decoder 270_1, a second column decoder 270_2, a third column decoder 270_3 and a fourth column decoder 270_4 and a first RMW control circuit 620_1.

[0160] The first row decoder 260_1 and the first column decoder 270_1 may be disposed adjacent to the first memory array 310_1. The second row decoder 260_2 and the second column decoder 270_2 may be disposed adjacent to the second memory array 310_2. The third row decoder 260_3 and the third column decoder 270_3 may be disposed adjacent to the third memory array 310_3. The fourth row decoder 260_4 and the fourth column decoder 270_4 may be disposed adjacent to the fourth memory array 310_4. The first RMW control circuit 620_1 may be disposed between the first memory array 310_1 to the fourth memory array 310_4.

[0161] The region where the first to fourth memory bank arrays 310_1 to 310_4 are disposed may correspond to the core region of the substrate. The region where the first to fourth row decoders 260_1 to 260_4, the first to fourth column decoders 270_1 to 270_4, and the first RMW control circuit 620_1 are disposed may correspond to the peripheral region of the substrate.

[0162] The first RMW control circuit 620_1 may include an oscillator 620_1a and a core signal generator 620_1b.

[0163] The oscillator 620_1a may generate an oscillation signal OSC in response to the internal activation count update signal IACU. As an example, the oscillator 620_1a may generate an oscillation signal OSC having a predetermined frequency in a portion where the internal activation count update signal IACU is at a high level. The frequency of the oscillation signal OSC may be set or variable in various ways according to the operating frequency of the memory device 200.

[0164] The core signal generator 620_1b may generate an RMW core signal group RMW_CSG for each memory bank based on the internal activation count update signal IACU and the oscillation signal OSC.

[0165] refer to Fig.12 and Fig.13A , the first RMW control circuit 620_1 may receive the first internal activation count update signal IACU1 from the IACU generator 610 at the zeroth time point t0. The first RMW control circuit 620_1 may generate an oscillation signal OSC based on the first internal activation count update signal IACU1. The oscillation signal OSC may be generated in a portion where the first internal activation count update signal IACU1 is at a high level.

[0166] The core signal generator 620_1 b may generate a plurality of intermediate signals A_R1 to A_R5 and A_F1 to A_F5 in synchronization with a rising edge or a falling edge of the oscillation signal OSC.

[0167] As an example, the core signal generator 620_1b may generate a first rising intermediate signal A_R1 in synchronization with a rising edge of the oscillation signal OSC at a first time point t1. The core signal generator 620_1b may generate a first falling intermediate signal A_F1 in synchronization with a falling edge of the oscillation signal OSC at a second time point t2. The core signal generator 620_1b may generate a second rising intermediate signal A_R2 in synchronization with a rising edge of the oscillation signal OSC at a third time point t3. The core signal generator 620_1b may generate a second falling intermediate signal A_F2 in synchronization with a falling edge of the oscillation signal OSC at a fourth time point t4. In this way, the core signal generator 620_1b may generate the first rising intermediate signal A_R1 to the fifth rising intermediate signal A_R5 and the first falling intermediate signal A_F1 to the fifth falling intermediate signal A_F5 at the first time point t1 to the tenth time point t10.

[0168] In addition, the core signal generator 620_1b may generate read control signals RD_CTRL1_1 and RD_CTRL_1_2 and write control signals WR_CTRL1_1 and WR_CTRL_1_2 to be applied to the first bank array 310_1, the first row decoder 260_1, and the first column decoder 270_1 based on the intermediate signals A_R1 to A_R5 and A_F1 to A_F5.

[0169] As an example, the core signal generator 620_1b may generate a first_first read control signal RD_CTRL1_1 based on the first rising intermediate signal A_R1 and the second rising intermediate signal A_R2. The core signal generator 620_1b may generate a first_second read control signal RD_CTRL1_2 based on the second rising intermediate signal A_R2 and the second falling intermediate signal A_F2. The core signal generator 620_1b may generate a first_first write control signal WR_CTRL1_1 based on the third rising intermediate signal A_R3 and the fourth falling intermediate signal A_F4. The core signal generator 620_1b may generate a first_second write control signal WR_CTRL_1_2 based on the third rising intermediate signal A_R3 and the fourth rising intermediate signal A_R4.

[0170] However, this is merely an example, and the read control signals RD_CTRL1_1 and RD_CTRL_1_2 and the write control signals WR_CTRL1_1 and WR_CTRL_1_2 may be generated in various ways, for example, may be generated based on other intermediate signals.

[0171] The read control signals RD_CTRL1_1 and RD_CTRL_1_2 and the write control signals WR_CTRL1_1 and WR_CTRL_1_2 generated by the core signal generator 620_1 b may constitute a first RMW core signal group RMW_CSG1 and may be used as control signals for controlling an RMW operation of the first memory bank Bank1 .

[0172] As described above, in case of receiving one internal activation count update signal IACU, the first RMW control circuit 620_1 may generate the RMW core signal group RMW_CSG based on the received internal activation count update signal IACU.

[0173] At the same time, as referenced Fig.10 and Fig. 11B As described above, the RMW driver 600 may receive first to fourth precharge commands PRE1 to PRE4. Based on the received first to fourth precharge commands PRE1 to PRE4, the IACU generator 610 of the RMW driver 600 may generate first to fourth internal activation count update signals IACU1 to IACU4 in parallel.

[0174] In this case, refer to Fig.12 and Fig. 13B , the first RMW control circuit 620_1 may sequentially receive the first to fourth internal activation count update signals IACU1 to IACU4. Based on the received first to fourth internal activation count update signals IACU1 to IACU4, the first RMW control circuit 620_1 may generate the first to fourth RMW core signal groups RMW_CSG1 to RMW_CSG4 in parallel.

[0175] As an example, the first RMW control circuit 620_1 may generate a first RMW core signal group RMW_CSG1 based on the first internal activation count update signal IACU1. The first RMW control circuit 620_1 may generate a second RMW core signal group RMW_CSG2 based on the second internal activation count update signal IACU2. In the same manner, the first RMW control circuit 620_1 may generate a third RMW core signal group RMW_CSG3 based on the third internal activation count update signal IACU3, and may generate a fourth RMW core signal group RMW_CSG4 based on the fourth internal activation count update signal IACU4.

[0176] In this case, at least some of the first to fourth RMW core signal groups RMW_CSG1 to RMW_CSG4 may overlap each other in the same period. In other words, at least some of the RMW operations for the first to fourth memory banks Bank 1 to Bank 4 may be performed substantially simultaneously.

[0177] As described above, even when the precharge commands PRE1 to PRE4 for different memory banks are continuously received at short time intervals, the RMW driver 600 according to the implementation of the present disclosure can generate the RMW control signal groups RMW_CSG1 to RMW_CSG4 for each memory bank in parallel. Therefore, the RMW operation of each memory bank can be performed substantially at the same time, and therefore, the RMW operation can be performed without reducing performance.

[0178] Fig.14A is a diagram showing an example of a memory bank array 310_1A. Fig.14A The memory bank array 310_1A may correspond to, for example, Figure 1 , Figure 5 , Figure 7 and Fig.12 one of the first memory bank arrays.

[0179] refer to Fig.14A , the memory array 310_1A may include a normal cell area NCA and a redundant cell area RCA.

[0180] The normal cell area NCA may include a first area RG1 storing normal data and a second area RG2 storing count data. The redundant cell area RCA may include a first area PRG1 storing parity data and a second area PRG2 storing count parity data. The parity data may be generated based on the normal data and may be used for an ECC encoding or decoding operation of the normal data. The count parity data may be generated based on the count data and may be used for an ECC encoding or decoding operation of the count data.

[0181] Fig. 14B is a diagram showing an example of a memory bank array 310_1B. Fig. 14B The bank array 310_1B may correspond to, for example Figure 1 , Figure 5 and Fig.12 one of the first memory bank arrays.

[0182] refer to Fig. 14B, the memory bank array 310_1B may include a plurality of sub-array blocks 311a to 315a. As an example, 311a to 314a of the sub-array blocks 311a to 315a may form a normal cell area NCA, and 315a of the sub-array blocks 311a to 315a may form a redundant cell area RCA.

[0183] At least some of the sub-array blocks 311a to 314a forming the normal cell area NCA may include a first region RG1 storing normal data and a second region RG2 storing count data. As an example, the first sub-array block 311a may include a first region RG1a storing normal data and a second region RG2a storing count data. The second sub-array block 312a may include a first region RG1b storing normal data and a second region RG2b storing count data. Meanwhile, in some embodiments, the fourth sub-array block 314a may be implemented to include only a first region RG1d storing normal data.

[0184] The sub-array block 315 a forming the redundant cell area RCA may include a first region PRG1 in which parity data is stored and a second region PRG2 in which count parity data is stored.

[0185] Fig.15 is a diagram showing an example of a memory bank array 310_C. Fig.15 The memory bank array 310_1C may correspond to, for example, Figure 1 , Figure 5 , Figure 7 and Fig.12 one of the first memory bank arrays.

[0186] refer to Fig.15 , the memory bank array 310_1C may include first sub-array blocks SCB11, 311a and 312a, second sub-array blocks SCB12, 313a and 314a, third sub-array blocks SCB2 and 315a, input / output sense amplifiers 331 to 335, and drivers 341 to 345.

[0187] Data input / output with respect to each of the first sub-array blocks 311a and 312a and the second sub-array blocks 313a and 314a may be performed through the first global input / output line GIO1<1:a> (a is a natural number) and the first local input / output line LIO1<1:a>. In response to a read command or a write command, a bit line may be selected by a column selection signal transmitted through a column selection line CSL in each of the first sub-array blocks 311a and 312a and the second sub-array blocks 313a and 314a.

[0188] The number of the first sub-array blocks 311 a and 312 a and the second sub-array blocks 313 a and 314 a arranged along the column direction may be determined in various ways according to the size of data bits processed by the memory device 200 .

[0189] Data input / output with respect to the third sub-array blocks SCB2 and 315a may be performed through the second global input / output lines GIO2<1:b> (b is a natural number less than a) and the second local input / output lines LIO2<1:b>. In response to a read command or a write command, a bit line may be selected by a column selection signal transmitted through the column selection lines CSLs in each third sub-array block 315a. The number of the third sub-array blocks 315a should not be limited to Fig.15 Quantity shown.

[0190] Similar to reference Fig. 14B As described above, the first sub-array blocks 311a and 312a may store normal data and count data, the second sub-array blocks 313a and 314a may store normal data, and the third sub-array block may store parity data and count parity data.

[0191] The input / output sense amplifiers 331 to 334 may correspond to the first and second sub-array blocks 311a, 312a, 313a, and 314a. Each of the input / output sense amplifiers 331 to 334 may sense and amplify a voltage output through the first global input / output line GIO1<1:a>.

[0192] The input / output sense amplifier 335 may correspond to the third sub-array block 315a and may sense and amplify a voltage of the second global input / output line GIO2<1:b> according to a bit output through the second global input / output line GIO2<1:b>.

[0193] The drivers 341 to 344 may correspond to the first and second sub-array blocks 311a, 312a, 313a, and 314a. The drivers 341 to 344 may transmit data to the memory cells of the first and second sub-array blocks 311a, 312a, 313a, and 314a through the first global input / output line GIO1<1:a> and the first local input / output line LIO1<1:a>.

[0194] The driver 345 may correspond to the third sub-array block 315a. The driver 345 may transmit data to memory cells of the third sub-array block 315a through the second global input / output line GIO2<1:b> and the second local input / output line LIO2<1:b>.

[0195] In some embodiments, during a read operation of an RMW operation, each of the input / output sense amplifiers 331 to 335 may operate in response to an input / output sense activation signal IOSA_EN. The input / output sense activation signal IOSA_EN may correspond to, for example, the first_second read control signal RD_CTRL1_2 (reference Figure 2A and Fig.13A ). In addition, a column selection signal that selects one of the column selection lines CSL during a read operation may correspond to a first_first read control signal RD_CTRL1_1 (reference Figure 2A and Fig.13A ).

[0196] In some embodiments, during a write operation of the RMW operation, each of the drivers 341 to 345 may operate in response to a write drive signal PDT. The write drive signal PDT may correspond to, for example, a first_first write control signal WR_CTRL1_1 (reference Figure 2A and Fig.13A ). In addition, the column selection signal that selects one of the column selection lines CSL during a write operation may correspond to the first_second write control signal WR_CTRL1_2 (reference Figure 2A and Fig.13A ).

[0197] As described above, the RMW operation of the first bank array 310_1C may be controlled based on the read control signals RD_CTRL1_1 and RD_CTRL_1_2 and the write control signals WR_CTRL1_1 and WR_CTRL_1_2 generated according to an embodiment of the present disclosure.

[0198] Fig.16 is a block diagram illustrating an example of a bank array 310_1D and a sense amplifier 285_1. Fig.16 The memory bank array 310_1D may correspond to, for example, Figure 1 , Figure 5 and Fig.12 one of the first memory bank arrays. Fig.16 The sense amplifier 285_1 may correspond to, for example, Figure 7 The first sense amplifier 285_1.

[0199] refer to Fig.16 , a plurality of sub-array blocks SCB may be disposed in the first memory bank array 310_1D. The sub-array blocks SCB may form J columns and I rows (each of I and J is a natural number).

[0200] A plurality of bit lines, a plurality of word lines, and memory cells located at points where the bit lines and the word lines intersect may be provided in each sub-array block SCB.

[0201] The sub word line driver region SWB may be disposed between the sub array blocks SCB. At least one sub word line driver may be disposed in each sub word line driver region SWB.

[0202] In addition, the bit line sense amplifier region BLSAB may be disposed between the sub-array blocks SCB. At least one bit line sense amplifier may be disposed in each bit line sense amplifier region BLSAB.

[0203] A junction area may be disposed adjacent to the sub word line driver area SWB and the bit line sense amplifier area BLSAB. As an example, a voltage generator may be disposed in the junction area.

[0204] The first sense amplifier 285_1 may include a plurality of input / output sense amplifiers (IOSA) 286a and a plurality of drivers (DRV) 287a. Each of the input / output sense amplifiers 286a may be connected to a corresponding sub-array block via corresponding global input / output lines GIO and GIOB. Similarly, each of the drivers 287a may be connected to a corresponding sub-array block via corresponding global input / output lines GIO and GIOB.

[0205] In some embodiments, the input / output sense activation signal IOSA_EN may be provided to each of the input / output sense amplifiers 286a during a read operation of the RMW operation. The input / output sense activation signal IOSA_EN may correspond to, for example, the first_second read control signal RD_CTRL1_2 (reference Figure 2A and Fig.13A ).

[0206] In some embodiments, a write drive signal PDT may be provided to each of the drivers 287a during a write operation of RMW. The write drive signal PDT may correspond to, for example, a first_first write control signal WR_CTRL1_1 (reference Figure 2A and Fig.13A ).

[0207] Fig.17 is shown in more detail Fig.16 FIG. 390 is an example of a portion of a first memory bank array. For ease of illustration, Fig.17 Only some word lines and some bit lines are shown.

[0208] refer to Fig.16 and Fig.17 , the sub array blocks SCBa and SCBb, the bit line sense amplifier region BLSAB, the sub word line driver region SWB, and the coupling region CONJ may be disposed in a portion 390 of the first memory bank array 310_1D.

[0209] The subarray block SCBa may include a plurality of word lines WL0 to WL3 extending in a row direction and a plurality of bit lines BL0 to BL3 extending in a column direction, and may include memory cells MC located at points where the word lines WL0 to WL3 intersect the bit lines BL0 to BL3 .

[0210] The sub-array block SCBb may include a plurality of word lines WL4 to WL7 extending in a row direction and a plurality of bit lines BL0 to BL3 extending in a column direction, and may include memory cells MC located at points where the word lines WL4 to WL7 intersect the bit lines BL0 to BL3 .

[0211] Sub word line driver regions SWBa1 and SWBa2 may include sub word line drivers 531, 532, 533 and 534 to drive word lines WL1, WL3, WL0, WL2, respectively. Sub word line driver regions SWBb1 and SWBb2 may include sub word line drivers 541, 542, 543 and 544 to drive word lines WL5, WL7, WL4 and WL6, respectively.

[0212] The bit line sense amplifier region BLSAB may include a bit line sense amplifier 650 connected to the bit line BL0 of the sub-array block SCBa and the bit line BL1 of the sub-array block SCBb in an open bit line structure, and a local sense amplifier circuit 680. The bit line sense amplifier 650 may amplify a level difference between voltages sensed at the bit lines BL0 and BL1, and may provide the amplified voltage level difference to the local input / output line pair LIO1 and LIOB1.

[0213] The coupling region CONJ may be located near the bit line sense amplifier region BLSAB, the sub word line driver region SWB, and the sub array block SCB. The voltage generators 510 and 520 may be disposed in the coupling region CONJ.

[0214] Fig.18 It is shown Fig.17 8 is a circuit diagram of an example of a bit line sense amplifier 650 .

[0215] refer to Fig.18 , the bit line sense amplifier 650 may be connected to the bit lines BL and BLB of each of the memory cells 660 and 670. The bit line sense amplifier 650 may include an N sense amplifier 651, a P sense amplifier 652, a precharge circuit 653, column selection switches 654a and 654b, an NSA driver 655, and a PSA driver 656.

[0216] The N sense amplifier 651 may discharge a low potential bit line among the bit lines BL and BLB during a sensing operation.

[0217] The N sense amplifier 651 may include NMOS transistors NM1 and NM2. The gate of the NMOS transistor NM1 may be connected to the bit line BLB, the drain of the NMOS transistor NM1 may be connected to the bit line BL, and the source of the NMOS transistor NM1 may be connected to the sense enable line LAB. The gate of the NMOS transistor NM2 may be connected to the bit line BL, the drain of the NMOS transistor NM2 may be connected to the bit line BLB, and the source of the NMOS transistor NM2 may be connected to the sense enable line LAB.

[0218] The N sense amplifier 651 may discharge the low potential bit line to the ground voltage VSS applied to the sense enable line LAB.

[0219] The P sense amplifier 652 may charge a high potential bit line among the bit lines BL and BLB with the level of the power supply voltage VDD during a sensing operation.

[0220] The P sense amplifier 652 may include PMOS transistors PM1 and PM2. The gate of the PMOS transistor PM1 may be connected to the bit line BLB, the source of the PMOS transistor PM1 may be connected to the bit line BL, and the drain of the PMOS transistor PM1 may be connected to the sense enable line LA. The gate of the PMOS transistor PM2 may be connected to the bit line BL, the source of the PMOS transistor PM2 may be connected to the bit line BLB, and the drain of the PMOS transistor PM2 may be connected to the sense enable line LA.

[0221] The P sense amplifier 652 may charge a high potential bit line among the bit lines BL and BLB using the power supply voltage VDD applied to the sense enable line LA.

[0222] In this case, the charging voltage VDD may be applied to the sensing enable line LA from the PSA driver 656. Then, the transistor PM2 whose gate is connected to the bit line BL whose voltage rises due to charge sharing may be turned off.

[0223] The precharge circuit 653 may precharge the bit lines BL and BLB to a level of half voltage (VDD / 2) in response to the control signal PEQ during the sensing operation. When the control signal PEQ is activated, the precharge circuit 653 may apply the bit line precharge voltage VBL to the bit lines BL and BLB for the sensing operation. The precharge circuit 653 may include NMOS transistors N3, N4, and N5.

[0224] The column selection switches 654a and 654b may connect the data sensed by the N sense amplifier 651 and the P sense amplifier 652 to the local input / output line pair LIO1 and LIOB1. As an example, when the sensing levels of the N sense amplifier 651 and the P sense amplifier 652 are stable during a read operation, the column selection line CSL may be activated. Then, the column selection switches 654a and 654b may be turned on, and the sensed data may be applied to the local input / output lines LIO1 and LIOB1.

[0225] The column selection switches 654 a and 654 b may include NMOS transistors N6 and N7 , respectively.

[0226] The NSA driver 655 may apply a driving signal to the sense enable line LAB of the N sense amplifier 651. The NSA driver 655 may receive a control signal LANG from the peripheral circuit 201. The NSA driver 655 may ground the sense enable line LAB in response to the control signal LANG. The NSA driver 655 may include a ground transistor N1 to control a voltage of the sense enable line LAB.

[0227] The PSA driver 656 may apply the charging voltage VDD to the sense enable line LA of the P sense amplifier 652. The PSA driver 656 may include a PMOS transistor P1. The PSA driver 656 may be controlled by a control signal LAPG. The control signal LAPG and the control signal LANG may have complementary logic levels.

[0228] In some embodiments, the column selection line CSL may be activated during a read operation of the RMW operation. In this case, the column selection signal that activates the column selection line CSL may correspond to the first_first read control signal RD_CTRL1_1 (reference Figure 2A and Fig.13A ).

[0229] In some embodiments, the column selection line CSL may be activated during a write operation of the RMW operation. In this case, the column selection signal activating the column selection line CSL may correspond to the first_second write control signal WR_CTRL1_2 (reference Figure 2A and Fig.13A ).

[0230] As reference Figures 16 to 18 As described above, the RMW operations of the first bank array 310_1D and the first sense amplifier 285_1 may be controlled based on the read control signals RD_CTRL1_1 and RD_CTRL_1_2 and the write control signals WR_CTRL1_1 and WR_CTRL_1_2 generated according to an embodiment of the present disclosure.

[0231] Fig.19 is a block diagram showing an example of a memory device that performs a normal read or write operation. For ease of explanation, assume that Fig.19 The memory bank array has Fig. 14B The first memory bank array has substantially the same configuration.

[0232] refer to Fig.19 , the I / O gating circuit 290 may include a plurality of switch circuits MUX respectively connected to the sub-array blocks 311 a to 315 a.

[0233] The ECC engine 350 may be connected to the switch circuit MUX corresponding to the sub-array blocks 311a to 314a via a data line (e.g., a global input / output line) corresponding to the normal cell area NCA. The ECC engine 350 may be connected to the switch circuit MUX corresponding to the sub-array block 315a via a data line corresponding to the redundant cell area RCA.

[0234] In the case of a normal write operation, the memory device 200 (refer to Figure 1 ) may receive a write command and data DTA. In this case, the ECC engine 350 may perform ECC encoding on the data DTA and may generate parity data PRT. Then, the ECC engine 350 may provide a codeword CW including the data DTA and the parity data PRT to the I / O gating circuit 290.

[0235] The codeword CW may be stored in one of the subarray blocks 311a to 314a of the normal cell area NCA. In this case, the codeword CW may be stored in one of the first regions RG1a to RG1d of the subarray blocks 311a to 314a. In addition, the parity data PRT may be stored in the subarray block 315a of the redundant cell area RCA. In this case, the parity data PRT may be stored in the first region PRG1 of the subarray block 315a.

[0236] In a normal read operation, the memory device 200 may receive a read command. In this case, the codeword CW stored in one of the subarray blocks 311a to 314a of the normal cell area NCA may be provided to the ECC engine 350 via the I / O gating circuit 290. In addition, the parity data PRT stored in the subarray block 315a of the redundant cell area RCA may be provided to the ECC engine 350 via the I / O gating circuit 290. The ECC engine 350 may decode the codeword CW based on the parity data PRT to generate data DTA.

[0237] Fig. 20 is a block diagram showing an example of a memory device that performs an RMW operation. For ease of explanation, assume that Fig. 20 The memory bank array has Fig. 14B In addition, it is assumed that a first precharge command PRE1 is received after a normal read operation or a write operation is performed on the first memory bank array.

[0238] When receiving the first precharge command PRE1, the RMW driver 600 may generate a first RMW core signal group RMW_CSG1 based on the first precharge command PRE1. As an example, the IACU generator 610 (refer to Fig.10 ) can generate a first internal activation count update signal IACU1 (reference Fig.10 ), and the first RMW control circuit 620_1 (reference Fig.10 ) may generate a first RMW core signal group RMW_CSG1 based on the first internal activation count update signal IACU1. The first RMW core signal group RMW_CSG1 may be provided to the I / O gating circuit 290 and the first memory bank array.

[0239] Then, a read operation of the RMW operation may be performed.

[0240] As an example, count data CNTD of each memory cell row may be stored in the second regions RG1a to RG1d of the subarray blocks 311a to 314a, and count parity data CPRT related to the count data CNTD may be stored in the second region PRG2 of the subarray block 315a.

[0241] In this case, count data CNTD corresponding to the target memory cell row may be read out from the second regions RG1a to RG1d of the sub-array blocks 311a to 314a, and the read count data CNTD may be provided to the ECC engine 350. In addition, count parity data corresponding to the target memory cell row may be read out from the second region PRG2 of the sub-array block 315a, and the read count parity data CPRT may be provided to the ECC engine 350. The ECC engine 350 may perform an ECC decoding operation on the count data CNTD and the count parity data CPRT to correct an error bit of the count data CNTD and may provide the count data CNTD to the row hammer management circuit 500.

[0242] Then, a correction operation of the RMW operation may be performed.

[0243] As an example, the row hammer management circuit 500 may perform an update operation to increase the count data CNTD by '1'.

[0244] Then, the write operation of the RMW operation can be performed.

[0245] As an example, the row hammer management circuit 500 may provide the updated count data UCNTD to the ECC engine 350. The ECC engine 350 may perform an ECC encoding operation on the updated count data UCNTD to generate updated count parity data UCPRT. The updated count data UCNTD may be stored again in the count cells corresponding to the activated memory cell row among the count cells of the second region RG2a to RG2c of the normal cell area NCA. The updated count parity data UCPRT may be stored again in the memory cells corresponding to the activated memory cell row among the memory cells of the second region PRG2 of the redundant cell area RCA.

[0246] As described above, since the memory device according to the present disclosure supports the RMW operation, the number of accesses to each memory cell row can be accurately managed, and thus the row hammer phenomenon caused by intensive accesses to a specific memory cell row can be prevented in advance.

[0247] In addition, the memory device according to the present disclosure can perform RMW operations based on the precharge command, and can perform RMW operations in parallel. Therefore, the RMW operation can be performed without degrading performance and increasing the complexity of the protocol between the host and the memory device 200.

[0248] at the same time, Figure 7 and Fig. 20 The memory device 200 shown in is described as including an ECC engine. Since the memory device 200 includes the ECC engine, the memory device 200 according to the present disclosure can accurately manage the count data of each storage cell row even when the activation count update time tACU is set to be relatively short, however, this is only an example. In some embodiments, the memory device 200 may not include the ECC engine.

[0249] Fig.21 is a flowchart illustrating an example of the operation of a memory device. Fig.21 The memory device described may be referenced Figures 1 to 20 One of the memory devices described.

[0250] In operation S110 , the memory device may receive an active command ACT from a host.

[0251] As an example, the memory device may include a plurality of memory banks and may receive an activation command ACT for each of the memory banks. In this case, a normal read or write operation may be performed on each memory bank.

[0252] In operation S120 , the memory device may continuously receive a precharge command PRE from the host.

[0253] As reference Figure 3 As described, the precharge command PRE sent from the host to the memory device may be sent earlier than the normal precharge command by the activation count update time tACU.

[0254] In operation S130 , the memory device may generate an internal activation count update signal IACU for each of the memory banks based on the precharge command PRE.

[0255] In this case, as referenced Fig. 13B As described, the memory device may generate an internal activation count update signal IACU for each memory bank. In addition, at least some of the activated segments corresponding to different internal activation count update signals IACU of different memory banks may overlap with each other.

[0256] In operation S140 , the memory device may generate an RMW core signal group RMW_CSG for each memory bank based on the internal activation count update signal IACU.

[0257] In operation S150 , the memory device may perform an RMW operation on each memory bank based on the RMW core signal group RMW_CSG of each memory bank.

[0258] In this case, the RMW operation of each memory bank can be performed in parallel. That is, the RMW operations for different memory banks can be performed substantially at the same time. Therefore, the RMW operation can be performed on each memory bank without reducing performance.

[0259] In operation S160 , a hammer refresh operation may be performed based on the count data or the updated count data.

[0260] As an example, in the case where the value of the updated count data generated by the RMW operation is greater than a predetermined value, the memory device may determine that the target memory cell row is an attacking memory cell row. In this case, the memory device may perform a hammer refresh operation on the adjacent memory cell row. Therefore, the row hammer phenomenon can be prevented in advance.

[0261] As described above, the memory device according to the present disclosure can manage the number of accesses to each of the memory cell rows through the RMW operation. Specifically, the memory device according to the present disclosure can perform the RMW operation based on the precharge command without defining a separate dedicated command for the RMW operation. Therefore, the RMW operation can be performed without reducing performance and increasing the complexity of the protocol between the memory controller 100 and the memory device 200.

[0262] At the same time, Figures 5 to 21In the embodiment, the RMW driver is described as being implemented independently of the control logic circuit, however, this is merely an example, and the present disclosure should not be limited thereto or thereby. As an example, part or all of the RMW driver may be implemented as being integrated with the control logic circuit. This will refer to Fig.22A , Fig. 22B , Fig.23A and Fig. 23B Describe in detail.

[0263] Fig.22A is a block diagram illustrating an example of a memory system 30 . Fig. 22B is shown in more detail Fig.22A 2 is a block diagram of an example of a control logic circuit 210. Fig.22A The memory system 30 is similar to Figure 5 The memory system 20. Fig.22A In the same reference numerals, Figure 5 The same elements are used herein, and thus detailed description of the same elements will be omitted.

[0264] with which the RMW driver and control logic circuits are implemented independently of each other Figure 5 Depending on the implementation, at least a portion of the RMW driver may be Fig.22A The control logic circuit 210 in is implemented in an integrated manner.

[0265] As an example, Fig.22A As shown, the IACU generator 610 may be implemented as a part of the control logic circuit 210 , and the RMW control circuit 620 may be implemented as a separate circuit independent of the control logic circuit 210 .

[0266] In this case, if Fig. 22B As shown, the control logic circuit 210 may include a command decoder 211, a mode register 212 and an IACU generator 610, and the control logic circuit 210 may generate an internal activation signal IACT, an internal write signal IWR, an internal read signal IRD, an internal pre-charge signal IPRE and an internal activation count update signal IACU based on a command CMD received from the memory controller 100.

[0267] Fig.23A is a block diagram illustrating an example of a memory system 30 . Fig. 23B is shown in more detail Fig.23A 2 is a block diagram of an example of a control logic circuit 210. Fig.23A The memory system 30 is similar to Figure 5 The memory system 20. Fig.23A In the same reference numerals, Figure 5 The same elements are used herein, and thus detailed description of the same elements will be omitted.

[0268] Fig.23A 2 shows a structure in which all components of the RMW driver are integrated with the control logic circuit 210. As an example, Fig.23A As shown, the IACU generator 610 and the RMW control circuit 620 may be implemented as part of the control logic circuit 210 .

[0269] In this case, if Fig. 23B As shown in FIG. 1 , the control logic circuit 210 may include a command decoder 211, a mode register 212, an IACU generator 610, and a plurality of RMW control circuits 620_1 to 620_n. The control logic circuit 210 may generate an internal activation signal IACT, an internal write signal IWR, an internal read signal IRD, and an internal precharge signal IPRE based on a command CMD received from the memory controller 100.

[0270] Furthermore, the control logic circuit 210 may generate a plurality of internal activation count update signals IACU1 to IACUn based on the precharge command PRE, and may generate an RMW core signal group RMW_CSG1 to RMW_CSGn for each memory bank based on the internal activation count update signals IACU1 to IACUn.

[0271] Although this specification contains many specific implementation details, these should not be interpreted as limitations on the scope of any invention or the scope that may be claimed, but rather as descriptions of features that may be specific to a particular implementation of a particular invention. Certain features described in this specification in the context of separate implementations may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single embodiment may also be implemented in multiples, either individually or in any suitable sub-combination. In addition, although features may be described above as working in certain combinations, one or more features from a combination may be deleted from the combination in some cases, and the combination may be directed to a sub-combination or a variation of a sub-combination.

[0272] While the present disclosure has been described with reference to the embodiments thereof, it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the present disclosure as set forth in the appended claims.

Claims

1. A memory device, include: A memory cell array comprising a plurality of memory cell rows, the memory cell array being configured to store count data of the number of accesses to each of the plurality of memory cell rows; The hammer management circuit is configured as follows: performing a read-modify-write (RMW) operation, the RMW operation reading out count data corresponding to a target memory cell row among the plurality of memory cell rows, Update the count data read out, and writing updated read count data into the memory cell array; and An RMW driver is configured to generate a plurality of RMW control signals, wherein the RMW driver is configured to generate the plurality of RMW control signals based on a precharge command and precharge the target memory cell row after a predetermined time has passed from a time point when the precharge command is applied. 2 . The memory device according to claim 1 , wherein the RMW operation for the target memory cell row and the precharge operation for the target memory cell row are performed based on the precharge command.

3. A memory device according to claim 2, wherein the RMW operation for the target memory cell row is performed based on the precharge command during an activation count update time, and the target memory cell row is precharged based on the precharge command after the activation count update time has passed from the time point when the precharge command is applied.

4. The memory device of claim 1 , wherein the memory cell array comprises a first memory bank and a second memory bank, the first memory bank and the second memory bank each comprising the plurality of memory cell rows, and wherein the RMW driver is configured as: receiving a first precharge command corresponding to the first memory bank, receiving a second precharge command corresponding to the second memory bank, generating a first internal activation count update signal corresponding to the first memory bank based on the first precharge command, and A second internal activation count update signal corresponding to the second memory bank is generated based on the second precharge command. 5 . The memory device of claim 4 , wherein a section in which the first internal activation count update signal corresponding to the first memory bank is activated overlaps a section in which the second internal activation count update signal corresponding to the second memory bank is activated.

6. The memory device of claim 4, wherein the RMW driver is configured to: generating a first RMW core signal group for controlling the RMW operation of the first memory bank based on the first internal activation count update signal, and A second RMW core signal group controlling the RMW operation of the second memory bank is generated based on the second internal activation count update signal.

7. The memory device of claim 1, wherein the RMW driver include: an internal activation count update signal generator configured to generate, based on the precharge command, an internal activation count update signal for a memory bank corresponding to the precharge command; and The RMW control circuit is configured to generate a RMW core signal group based on the internal activation count update signal, wherein the RMW core signal group controls the RMW operation of the memory bank corresponding to the precharge command. 8 . The memory device of claim 7 , wherein the internal activation count update signal generator is configured to maintain a level of the internal activation count update signal at a high level during the predetermined time.

9. The memory device of claim 8, wherein the RMW control circuit include: an oscillator configured to generate an oscillation signal during the high level of the internal activation count update signal; and A core signal generator is configured to generate the RMW core signal group based on a rising edge or a falling edge of the oscillation signal, wherein the RMW core signal group controls the RMW operation of the memory bank corresponding to the precharge command.

10. The memory device of claim 9, wherein the oscillation signal has a frequency that changes as an operating frequency of the memory device changes.

11. The memory device according to claim 1 , comprising a control logic circuit, wherein the control logic circuit is configured to generate an internal precharge signal based on the precharge command, and the internal precharge signal precharges the target memory cell row after the predetermined time has passed from the time point when the precharge command is applied. 12 . The memory device of claim 1 , wherein the row hammer management circuit is configured to perform a refresh operation on a memory cell row adjacent to the target memory cell row based on a value of the updated sensed count data being equal to or greater than a predetermined value.

13. The memory device of claim 1, comprising an ECC engine configured to perform an error correction operation on the read count data.

14. A memory device, include: A memory cell array comprises a first memory bank and a second memory bank, wherein the first memory bank and the second memory bank each comprise a plurality of memory cell rows; a row hammer management circuit configured to perform a first read-modify-write (RMW) operation on the first memory bank and a second RMW operation on the second memory bank in parallel; and An RMW driver is configured to generate a first RMW core signal group and a second RMW core signal group in parallel, the first RMW core signal group controlling the first RMW operation for the first storage body, and the second RMW core signal group controlling the second RMW operation for the second storage body, wherein the first RMW core signal group is generated based on a first precharge command for the first storage body, and the second RMW core signal group is generated based on a second precharge command for the second storage body, a first precharge operation corresponding to the first precharge command is performed after a time delay of an activation count update time from a time point when the first precharge command is applied, and a second precharge operation corresponding to the second precharge command is performed after a time delay of an activation count update time from a time point when the second precharge command is applied.

15. The memory device according to claim 14 comprises a control logic circuit, which is configured to generate a first internal precharge signal and a second internal precharge signal based on the first precharge command and the second precharge command, respectively, wherein the first internal precharge signal is generated after the time delay of the activation count update time from the time point of applying the first precharge command, and the second internal precharge signal is generated after the time delay of the activation count update time from the time point of applying the second precharge command.

16. The memory device according to claim 15, in, The RMW driver comprises an internal activation count update signal generator, the internal activation count update signal generator being configured to: generating a first internal activation count update signal corresponding to the first memory bank and a second internal activation count update signal corresponding to the second memory bank, changing the level of the first internal activation count update signal from a low level to a high level based on the first precharge command, transitioning the level of the first internal activation count update signal from the high level to the low level based on the first internal precharge signal, transitioning the level of the second internal activation count update signal from the low level to the high level based on the second precharge command, and The level of the second internal activation count update signal is transitioned from the high level to the low level based on the second internal precharge signal.

17. The memory device of claim 16, wherein the RMW driver comprises a RMW control circuit, the RMW control circuit being configured to: generating the first RMW core signal group corresponding to the first memory bank based on the first internal activation count update signal, and generating the second RMW core signal group corresponding to the second memory bank based on the second internal activation count update signal, and in, The RMW control circuit comprises: oscillator, which is configured as: generating a first oscillation signal during a time when the first internal activation count update signal is maintained at the high level, and generating a second oscillation signal during a time when the second internal activation count update signal is maintained at the high level; and The core signal generator is configured to: generating the first RMW core signal group based on a rising edge or a falling edge of the first oscillation signal, and The second RMW core signal group is generated based on a rising edge or a falling edge of the second oscillation signal.

18. A method of operating a memory device, include: receiving an activation command corresponding to a target memory cell row among a plurality of memory cell rows; receiving a precharge command corresponding to the target memory cell row; generating an internal activation count update signal corresponding to the target memory cell row based on the precharge command; generating a plurality of control signals based on the internal activation count update signal, the plurality of control signals updating count data for the target memory cell row; and After the activation count update time has elapsed, a precharge operation is performed on the target memory cell row based on the precharge command.

19. The method of claim 18, comprising performing a read-modify-write (RMW) operation, wherein the RMW operation reads count data corresponding to the target memory cell row during the activation count update time, updates the read count data, and writes the updated read count data into the memory cell array. 20 . The method of claim 18 , comprising performing a refresh operation on a memory cell row adjacent to the target memory cell row based on a value of the count data for the target memory cell row being equal to or greater than a predetermined value.

Citation Information

Patent Citations

  • Method for detecting manipulated or falsified GNSS signals

    KR1020230071163A

  • Compression pipe connector

    KR1020230132061A