WO2017218227A1 - Shared error detection and correction memory - Google Patents

Shared error detection and correction memory Download PDF

Info

Publication number
WO2017218227A1
WO2017218227A1 PCT/US2017/035946 US2017035946W WO2017218227A1 WO 2017218227 A1 WO2017218227 A1 WO 2017218227A1 US 2017035946 W US2017035946 W US 2017035946W WO 2017218227 A1 WO2017218227 A1 WO 2017218227A1
Authority
WO
WIPO (PCT)
Prior art keywords
test
memory
address information
circuit
defective
Prior art date
Application number
PCT/US2017/035946
Other languages
French (fr)
Inventor
Tomoyuki Shibata
Chikara Kondo
Hiroyuki Tanaka
Original Assignee
Micron Technology, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Micron Technology, Inc. filed Critical Micron Technology, Inc.
Priority to KR1020207030188A priority Critical patent/KR102399014B1/en
Priority to CN202210673440.8A priority patent/CN114974393A/en
Priority to KR1020217030505A priority patent/KR20210123403A/en
Priority to EP17813792.3A priority patent/EP3472708A4/en
Priority to CN201780037240.0A priority patent/CN109313596B/en
Priority to KR1020187036759A priority patent/KR102170322B1/en
Publication of WO2017218227A1 publication Critical patent/WO2017218227A1/en

Links

Classifications

    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/38—Response verification devices
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16—Error detection or correction of the data by redundancy in hardware
    • G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2017—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where memory access, memory control or I/O control functionality is redundant
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/1201—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details comprising I/O circuitry
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/14—Implementation of control logic, e.g. test mode decoders
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/36—Data generation devices, e.g. data inverters
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/38—Response verification devices
    • G11C29/42—Response verification devices using error correcting codes [ECC] or parity check
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/44—Indication or identification of errors, e.g. for repair
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/44—Indication or identification of errors, e.g. for repair
    • G11C29/4401—Indication or identification of errors, e.g. for repair for self repair
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/48—Arrangements in static stores specially adapted for testing by means external to the store, e.g. using direct memory access [DMA] or using auxiliary access paths
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/70—Masking faults in memories by using spares or by reconfiguring
    • G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/70—Masking faults in memories by using spares or by reconfiguring
    • G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
    • G11C29/84—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved access time or stability
    • G11C29/846—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved access time or stability by choosing redundant lines at an output stage
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C2029/0407—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals on power on
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C2029/1208—Error catch memory
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/36—Data generation devices, e.g. data inverters
    • G11C2029/3602—Pattern generator
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C2029/4402—Internal storage of test result, quality data, chip identification, repair information
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C5/00—Details of stores covered by group G11C11/00
    • G11C5/02—Disposition of storage elements, e.g. in the form of a matrix array
    • G11C5/025—Geometric lay-out considerations of storage- and peripheral-blocks in a semiconductor storage device

Definitions

  • 3D memory devices have been introduced. Some 3D memory devices are formed by stacking chips (e.g., dice) vertically and interconnecting the chips using through substrate vias (TSVs). Benefits of the 3D memory devices include shorter interconnects which reduce circuit delays and power consumption, a large number of vertical vias between layers which allow wide bandwidth buses between functional blocks in different layers, and a considerably smaller footprint Thus, the 3D memory devices contribute to higher memory access speed, lower power consumption and chip size reduction.
  • Example 3D memory devices include Hybrid Memory Cube (HMC), High Bandwidth Memory (HBM), and a wide-I/O dynamic random access memory (DRAM).
  • HMC Hybrid Memory Cube
  • HBM High Bandwidth Memory
  • DRAM wide-I/O dynamic random access memory
  • High Bandwidth Memory is a type of memory including a high- performance DRAM interface and vertically stacked DRAM.
  • a typical HBM stack of four DRAM chips e.g., core chips
  • I/F interface
  • An interface (I/F) chip of the HBM provides an interface with the eight input/output channels, which function independently of each other. For example, a clock frequency, a command sequence, and data can be independently provided for each channel.
  • the eight input/output channels are not necessarily synchronous to each other.
  • a type of test can be performed using a memory Built-in Self Test (mBIST) circuit that may be provided on the I/F chip.
  • the mBIST circuit is provided for verifying failures resulting from stacking the chip.
  • the mBIST circuit may include a memory to store defect information called an error-catch memory (ECM).
  • ECM error-catch memory
  • the HBM has a post package repair function performed by using the mBIST circuit
  • the post package repair function uses redundancy cells for repair and these redundancy cells are normally formed in memory matrices of the core.
  • the post package repair function may not be able to repair defects if the number of defective cells is greater than a number of repairable cells by providing redundancy cells.
  • it may be difficult to repair one or more defective cells which are redundancy cells.
  • An example apparatus may include at least one memory chip and an interface chip.
  • the at least one memory chip may include a plurality of first memory cells.
  • interface chip may be coupled to the at least one memory chip and may include a control circuit and a storage area.
  • the control circuit may detect one or more defective memory cells of the first memory cells of the at least one memory chip and may further store first defective address information of the one or more defective memory cells of the first memory cells into the storage area.
  • the interface chip may respond, at least in part, to the first defective address information and an access request to access the storage area in place of the at least one memory chip when the access request has been provided with respect to the one or more defective memory cells of the first memory cells.
  • An example interface chip apparatus may include a test circuit including a test logic circuit and one or more memories.
  • the test logic circuit may detect one or more defective memory cells of at least one memory chip when the test circuit is coupled to the at least one memory chip.
  • a first portion of the one or more memories may store first address information of the one or more defective memory cells as defective address information.
  • the test circuit may access one or more memory cells in a second portion of the one or more memories responsive to an access request, when the test logic circuit determines that the access request includes second address information relevant to the one or more defective memory cells based on the defective address information.
  • An example method of sharing error correction memory on an interface chip may include detecting one or more defective memory cells of a memory chip; storing first address information of the one or more defective memory cells as defective address information; and accessing one or more memory cells in in the error correction memory responsive to an access request including second address information relevant to the one or more defective memory cells based on the defective address information.
  • Fig. 1A is a schematic diagram of an interface (I/F) chip and a plurality of core chips in a semiconductor device in accordance with an embodiment of the present disclosure.
  • Fig. IB is a schematic diagram of a memory system including the semiconductor device that includes the interface (I/F) chip and the plurality of core chips in accordance with an embodiment of the present disclosure.
  • Fig. 2 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure.
  • Fig. 3 is a block diagram of the interface circuit in Fig. 2 including a memory Built-in
  • mBIST Self Test
  • Fig. 4A is a simplified flow diagram of a test operation process by the mBIST in Fig.
  • Fig. 4B is a simplified flow diagram of a test operation as a part of an initialization operation process performed by the mBIST in the semiconductor device in accordance with an embodiment of the present disclosure.
  • FIG. 5 is a schematic diagram of the data input/output circuit of Fig. 3, in accordance with an embodiment of the present disclosure.
  • FIG. 6 is a schematic diagram of the access signal output circuit of Fig. 3, in accordance with an embodiment of the present disclosure.
  • Fig. 7A is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
  • Fig. 7B is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
  • Fig. 7C is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
  • Fig. 7D is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
  • Fig. 8 is a block diagram of the I/F chip including a memory Built-in Self Test
  • Fig. 9 is a schematic diagram of one core chip of the plurality of core chips in the semiconductor device in accordance with an embodiment of the present disclosure.
  • Fig. 10 is a block diagram of the I/F chip including a memory Built-in Self Test (mBIST) circuit in the semiconductor device in accordance with an embodiment of the present disclosure.
  • mBIST memory Built-in Self Test
  • Fig. 1 1A is a simplified flow diagram of a write operation in the I/F chip of Fig. 10, in accordance with an embodiment of the present disclosure.
  • Fig. 1 IB is a simplified flow diagram of a read operation in the I/F chip of Fig. 10, in accordance with an embodiment of the present disclosure.
  • Fig. 12 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure.
  • Fig. 13 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure.
  • Fig. 14A is a simplified flow diagram of a write operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure.
  • Fig. 14B is a simplified flow diagram of a read operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure.
  • Fig. 1A is a schematic diagram of an interface (I/F) chip and a plurality of core chips in a semiconductor device in accordance with an embodiment of the present disclosure.
  • the semiconductor device 10 may be a 3D memory device, such as an HBM, an HMC, a Wide-IO DRAM etc.
  • the semiconductor device is formed by stacking chips vertically, as shown in Fig. 1A.
  • the stacked chips may include an interface chip 21 and core chips 22.
  • each core chip 22 may be a memory chip which includes two channels. Each channel may include a plurality of memory cells and circuitries accessing the memory cells.
  • the memory cells may be DRAM memory cells.
  • Fig. IB is a schematic diagram of a memory system including the semiconductor device that includes the interface (I/F) chip and the plurality of core chips in accordance with an embodiment of the present disclosure.
  • the memory system 1 may include a memory controller 11 and the 3D memory device 10.
  • the core chips 22a, 22b, 22c and 22d include channels A and C, channels B and D, channels E and G, and channels F and H, respectively.
  • the channels A, B, C, D, E, F, G and H of the core chips 22 may be coupled to the I/F chip 21 via different signal lines 23a, 23 b, 23c, 23d, 23e, 23f, 23g, and 23h, respectively.
  • the I/F chip 21 may include test control terminals 24, data terminals 25 and access signal terminals 26 which receive corresponding signals from a memory controller 11.
  • the test control terminals 24 may receive test control signals.
  • the data terminals 25 may receive write data from the memory controller 11 or may transmit read data to the memory controller 1 1.
  • the access signal terminals 26 may receive access signals from the memory controller 11.
  • the access signals may include an operation command (e.g., a read command, a write command) and address information corresponding to the operation command.
  • Fig. 2 is a block diagram of an interface circuit on the I/F chip 21 in the semiconductor device in accordance with an embodiment of the present disclosure.
  • An interface circuit 31 is provided on the I/F chip 21.
  • the interface circuit 31 may include channel interface blocks CIF_A 32a, CEF_B 32b, ... and CIF_H 32h which are coupled to corresponding channels, channel A, channel B and channel H of the core chip 22 in Fig.
  • the interface circuit 31 may include a plurality of data through substrate vias, TSVDs 33 and access through substrate vias TSVAs 34.
  • the TSVDs 33 and TSVAs 34 are configured to couple the channel interface blocks CIF_A 32a, CIFJB 32b, ... and CIF_H
  • each signal line 23 in Fig. IB may include the corresponding TSVD 33 and TSVA 34 for each channel.
  • Each channel interface block of the channel interface blocks CIF_A 32a, CIF_B 32b, ... and CIF_H 32h may include a data input/output circuit DI/O 35 and an access signal output circuit ASO 36.
  • Each DI/O 35 may receive data from the coupled channel via the corresponding TSVD 33.
  • the DI/O 35 may be further coupled to data terminals DT 37.
  • the data terminals DT 37 may be the data terminals 25 in Fig. IB.
  • the ASO 36 may be coupled to access signal terminals AT 38.
  • the access signal terminals AT 38 may receive access signals including access requests provided externally (e.g., from the memory controller 11 of Fig. IB). Each access request may include a command including an operation requested (e.g., a read request, a write request, etc.) and an address where the operation requested by the command is operated.
  • the access signal terminals AT 38 may be the access signal terminals 26 in Fig. IB.
  • Each ASO 36 may provide the access requests as command signals and address signals, for example, to the coupled channel via the corresponding TSVA 34. In a normal read operation, a read request and corresponding address information are externally provided to the ASO 36 via the AT 38.
  • the ASO 36 may provide the read request and the corresponding address information to the coupled channel via the TSVA 34, responsive to the read request
  • the coupled channel may provide the data to the DI/O 35 via the TSVD 33 in response to the read request and the corresponding address information.
  • the DI/O 35 may provide the data via the DT 37.
  • a write request and corresponding address information are externally provided to the ASO 36 via the AT 38.
  • the ASO 36 may provide the write request and the corresponding address information to the coupled channel via the TSVA 34, responsive to the write request.
  • the DI/O 35 may receive write data from the DT 37 and provide the write data to the coupled channel via the TSVD 33 responsive to the write request and the corresponding address information.
  • Each of the channel interface blocks CIF_A 32a, CIF_B 32b, ... and CIF_H 32h may further include a data comparator circuit DCMP 39 and an access signal comparator circuit ACMP 40. Functionalities of the DCMP 39 and the ACMP 40 will be provided later in detail.
  • the interface circuit 31 may further include a memory built-in self-test block mBIST 41.
  • the mBIST 41 may provide test signals to the channel interface blocks CIF_A 32a, CEF B 32b, ... and CIF H 32h via a multiplexer MUX 45 in order to perform test operations on each channel responsive to test control signals TCTL.
  • the TCTL are externally provided via test control terminals TCT 43.
  • the TCT 43 may be the test control terminals 24 in Fig. IB.
  • the mBIST 41 may perform defective cell detection and repair functions during an initialization operation and a normal operation following the initialization operation, such as read and write operations, as well as during a test operation. Functionalities of the mBIST 41 will be provided later in detail.
  • the interlace circuit 31 may further include a power detection circuit PD 42.
  • the PD 42 may provide a power on signal PON to the mBIST 41 responsive to power supply PS externally provided via a power terminal PT 44.
  • the mBIST 41 may start the initialization operation in response to the PON.
  • the interface circuit 31 may further include test terminals TT 46 and test pads TP 47.
  • the TT 46 may be coupled to a socket or the like in order to couple an external tester (not shown) to the interface circuit 31.
  • the TP 47 may couple an external tester (not shown) having a probe card interface, for example, to the interface circuit 31.
  • the TP 47 may be used for a test during wafer process.
  • the MUX 45 may select one of the TT 46, the TP 47 and the mBIST 41 as a source of the test signals and provide the test signals to the channel interface blocks CDF_A 32a, CIFJB 32b, ... and CIF_H 32h.
  • the terminals such as the DT 37, the AT 38, the TCT 43, the PT 44 and the
  • TT 46 represented by circles may be formed by micro bump electrodes, for example.
  • the test pad TP 47 may be formed by a pad electrode, for example.
  • Each of the DT 37, the AT 38, the TCT 43, the PT 44 and the TT 46 may include one or more terminals.
  • the TP 47 may include one or more pads.
  • Fig. 3 is a block diagram of the interface circuit 31 in Fig. 2 including the memory Built-in Self Test (mBIST) circuit 41 in the semiconductor device in accordance with an embodiment of the present disclosure.
  • Fig. 3 shows a connection between mBIST 41 of Fig. 2 and one of the channel interface blocks CIF_A 32a, CIF_B 32b, ... and CIF_H 32h of Fig. 2 as an example.
  • the other channel interface blocks may be coupled to the mBIST 41 similarly as shown in Fig. 3.
  • the interface chip may include a plurality of TSVD[n:0] 33 and corresponding DI/O[n:0] 35, DT[n:0] 37 and DCMP[n:0] 39provided with a single channel.
  • the interface chip may include a plurality of TSVA[m:0] 34, ASO[m:0] 36, AT[m:0] 38 and ACMP[m:0] 40provided with a single channel.
  • m is a number of bits for command/address signals
  • n is a number of bits for data signals
  • M m" and "n” may be different from each other.
  • the mBIST 41 may include an mBIST logic circuit mBISTL 411 and a storage area MEM 412.
  • the mBISTL 411 may be a control circuit, such as an algorithmic pattern generator (APG), which may control operations of the mBIST 41.
  • the MEM 412 may be a single memory circuit.
  • the MEM 412 may be a plurality of memory circuits, each of which may perform independent functions individually.
  • the MEM 412 may include error catch memories (ECMs) and microcode memories (MCMs).
  • ECMs error catch memories
  • MCMs microcode memories
  • each of the ECMs and the MCMs may include static random access memories (SRAMs).
  • the MCMs may store microcodes.
  • the microcodes may represent test patterns for testing the memory cells of the core chips.
  • the mBISTL 411 may perform test operations in accordance with the microcodes.
  • the ECMs may store defective address information during the test operations.
  • the MEM 412 such as the ECMs and/or the MCMs, can be used as spare cells for repairing defective memory cells in the core chips 22 as described later in detail.
  • the I/F chip 21 may further include a read only memory (ROM) to store the microcodes.
  • initial states of the MCMs in the MEM 412 may represent the microcodes.
  • the mBIST 41 may operate during the test operations.
  • the mBIST 41 may further operate during the initialization operation and the normal operation.
  • the mBIST 41 may detect one or more defective cells among memory cells of core chips 22.
  • the mBIST 41 may store address information for the defective cells in one portion of the MEM 412, for example the ECMs and/or MCMs, during the initialization operation.
  • the mBIST 41 may use an other portion of the MEM 412, for example, the ECMs and/or MCMs, to replace the defective cells with the other portion of the MEM 412 which may serve as spare memory.
  • the mBIST 41 may redirect access to the defective cells of the core chips 22 to the other portion of the MEM 412.
  • each of the core chips 22 may further include a defective address storing circuit, such as anti-fuses and spare memory cells.
  • a defective address storing circuit such as anti-fuses and spare memory cells.
  • the other portion of the MEM 412 of the I/F chip 21 and the spare memory cells of the core chips 22 may be used for repairing different defective cells from each other.
  • the other portion of the MEM 412 of the I/F chip 21 may replace the defective cells of the spare memory cells of the core chips 22.
  • Fig. 4A is a simplified flow diagram of a test operation process by the mBIST in Fig.
  • test operation process may be performed by the mBIST before shipping.
  • the mBIST 41 may receive the test control signals TCTL externally provided through the test control terminals TCT 43.
  • the test control signals may include a test instruction.
  • the mBISTL 411 may execute the microcodes to perform a test operation.
  • the microcodes may be prestored on the MCMs or a ROM of the I/F chip 21 as their initial states.
  • the MCMs may store the microcodes externally provided.
  • the mBISTL 411 may provide a test enable signal TEST to the DI/O[n:0] 35 and the ASO[n:0] 36.
  • the DI/O[n:0] 35 may couple, in response to the TEST, the TSVD[n:0] 33 to a BISTDATA node coupled to the DCMP[n:0] 39 and the mBIST 41 while decoupling the external data terminal DT[n:0] 37 from the DCMP[n:0] 39 and the mBIST 41.
  • the ASO[m:0] 36 may couple, in response to the TEST, the TSVA[m:0] 34 to a BIST command and address node BISTC/A coupled to the ACMP[m:0] 40 and the mBIST 41 while decoupling the external access terminal AT[m:0] 38 from the ACMP[m:0] 40 and the mBIST 41.
  • the detailed description of the DI/O[n:0] 35 and the ASO[m:0] 36 will be provided later referring to Figs. 5 and 6.
  • the mBISTL 411 may provide test write data
  • the mBISTL 411 may provide a test command/address signal TC/A[m:0] including test write commands and test address information to the ASO[m:0] 36. Accordingly, the DI/O[n:0] 35 and the ASO[m:0] 36 may perform test write operations on the respective channel. For example, the DI/O[n:0] may provide the TWDATA[n:0J as core data COREDATA[n:0] to the TSVD[n:0] 33.
  • the ASO[m:0] 36 may provide the TC/A[m:0] as a core command and address signal COREC/A[m:0] to the TSVA[m:0] 34.
  • the mBISTL 411 may provide the test command/address signal TC/A[m:0] test including read commands and test address information to the ASO[m:0] 36 to perform test read operations in an operation block 403.
  • the mBISTL 411 may provide a memory control signal MEMCTL including the test address information TA to the MEM 412.
  • the ASO[m:0] 36 may provide the COREC/Aim:01 including the test read commands and the test address information to the respective channel via the TSVA[m:0] 34.
  • the respective channel may provide the COREDATA[n:0] including test read data to the TSVD[n:0] 33 responsive to the COREC/A[m:0].
  • the DI/O[n:0] 35 may receive the test read data and provide the test read data to the DCMP[n:0] 39.
  • the DCMP[n:0] 39 may compare the test read data with the expected data EXP[n:0] that is the corresponding test write data provided from the mBISTL 411.
  • the DCMP[n:0] 39 may provide a comparison result DCMR[n:0] to an error detection circuit PFD 48. If at least one DCMR[n:0] is indicative of data discrepancy between the test read data and the EXP[n:0], the PFD 48 may provide an active fail detection signal P/F (e.g., at a logic high level) in an operation block 406.
  • the PFD 48 may provide an inactive P/F (e.g., at a logic low level).
  • the MEM 412 may store the corresponding test address information as defective address information in an operation block 406.
  • determined defective address information stored in the MEM 412 may be programed into a defective address storing circuit, for example anti-fuses, of the core chips 22 in an operation block 407.
  • the determined defective address information may be read out from the MEM 412 to an external tester or an external controller (not shown) before programming in an operation block 40S.
  • the programing of the defective address information may be performed by mBISTL 411 or externally by the tester or the controller.
  • Fig. 4B is a simplified flow diagram of a test operation as a part of an initialization operation process performed by the mBIST in the semiconductor device in accordance with an embodiment of the present disclosure.
  • the test operation as the part of the initialization operation process is similar to the test operation before shipping described above except a power on sequence.
  • the mBISTL 411 may respond to a power on signal PON instead of the TCTL in order to initiate testing of the memory cells.
  • the TCTL may be used to initiate testing in an operational block 40 F that is similar to the operation block 401 in Fig. 4A.
  • the programming of the determined defective address information into the defective address storing circuit of the core chips 22 may be skipped.
  • the determined defective address information may be stored in the MEM 412 of the I/F chip 21 in an operation block 406', after executing the test write in an operation block 402', similar to the operation block 402, the test read in an operation block 403' similar to the operation block 403 and the comparison of the test read data and the expected data EXP[n:0] in an operation block 404' that is similar to the operation block 404.
  • the normal access operation including a write operation and a read operation, may follow the initialization operation.
  • the TEST may be inactive (e.g., at a logic low level).
  • the DI/O[n:0] 35 may couple the TSVD[n:0] 33 to the DT[n:0] 37 until hit signals Hit_R/W become active (e.g., at a logic high level).
  • Hit_R/W become active (e.g., at a logic high level).
  • An external memory controller e.g., the memory controller 11 of Fig. IB
  • the external memory controller may also provide an EXC/A[m:0] signal including a write command and access address information (e.g., write address information) to the AT[m:0] 38.
  • the ASO[m:0] 36 may provide the EXC/A[m:0] as the COREC/A[m:0] to the respective channel.
  • the ASO[m:0] 36 may further provide the EXC/A[m:0] on the BISTC/A node to the ACMP[m:0] 40.
  • the ACMP[m:0] 40 may compare the access address information on the BISTC/A node with the defective address information DEFAIF[m:0] provided from the MEM 412.
  • the ACMP[m:0] 40 may provide a comparison result ACMRfmiO] to a hit detection circuit HITD 49.
  • the HITD 49 may provide a HIT signal to the mBISTL 411.
  • the MEM 412 may provide the DEFAIF[m:0] to the ACMP[m:0] 40.
  • the ACMP[m:0] 40 may include latch circuits, for example, which may hold the DEFAIF[m:0] at least during the normal operation (e.g., the write operation or the read operation).
  • the HIT signal may indicate that the provided address information corresponds to the defective address information.
  • the HIT signal may also indicate a type of the operation (e.g., the write operation).
  • the mBIST 41 may provide a HIT W signal to the DI/O[n:0] 35. Responsive to the HIT_W signal, the DI/O[n:0] 35 provides the EXDATA[n:0] on the BISTDATA node to the MEM 412.
  • the mBISTL 411 may perform the write operation on the MEM 412 to store the write date into the MEM. In some embodiments, the mBISTL 411 may stop performing a write operation on the respective channel on the core chip 22 while the repair operation, such as redirecting the access to the MEM 412, is being performed.
  • the DI/O[n:0] 35 may provide the EXDATA[n:0] to the channel as the COREDATA[n:0] and the COREDATA[n:0] are written into the corresponding memory cells of the core chip 22.
  • the normal read operation is similar to the normal write operation described above except a reading sequence.
  • the mBISTL 411 may provide the HIT R signal in response to the HIT signal provided from the HITD 49 when any ACMR[m:0] is indicative of a match between the access address information (e.g., read address information) and the DEFAJLF[m:0].
  • the DI/O[n:0] 35 may couple the DT[n:0] 37 to the MEM 412 through die BISTDATA node.
  • the mBISTL may further perform the read operation on the MEM 412 to provide requested read data from the MEM 412.
  • the mBISTL 41 1 may control a read latency for the read data.
  • the read data retrieved from the MEM 412 may be provided to the DI/O[n:0] 35 via the BISTDATA node.
  • the DI/O[n:0] 35 may selectively couple the BISTDATA node to the EXDATA node responsive to the active HIT_R signal, thus the DI/O[n:0] 35 may provide the read data to the DT[n:0] 37.
  • the mBISTL 411 may stop performing a read operation on the respective channel on the core chip 22 while the repair operation, such as redirecting the access to the MEM 412 for reading, is being performed.
  • the address information on the BISTC/A node does not correspond to the DEFAJF[m:0]
  • the data may be retrieved from the memory cells of the respective channel of core chip 22 to the DT[n:0] 37.
  • Fig. 5 is a schematic diagram of the data input/output circuit (DI/O) 35 of Fig. 3, in accordance with an embodiment of the present disclosure.
  • the DI/O 35 is coupled to the EXDATA node.
  • the DI/O 35 may include a receiver buffer Rx 50.
  • the receiver buffer Rx 50 receives the EXDATA from the DT 37 and may further provide the EXDATA to one input node of a selector 55.
  • the selector 55 may further receive BISTDATAout on the BISTDATA node from the DCMP 39 at another input node.
  • the selector 55 may further receive the TEST from the mBIST 41 at a select node after inversion.
  • the TEST may be active in the test operation and inactive in the normal operation (e.g., a write operation, a read operation, etc.). If the TEST is inactive, the selector 55 may provide the EXDATA to a buffer circuit 57. If the TEST is active, the selector 55 may provide the BISTDATAout from the BISTDATA node to the buffer circuit 57.
  • the buffer circuit 57 may be a tri- state buffer. An input node of the buffer circuit 57 may receive a signal from the selector 55. An enable input node of the buffer circuit 57 may receive a write enable signal CORE W, indicative of enabling writing data to the core chips 22.
  • the CORE_W may be provided as a portion of the COREC/A by the ASO 36 in response to a write command.
  • the buffer circuit 57 may provide the signal from the selector 55 as the COREDATA when the CORE_W is active (e.g., a logic high level).
  • the DI/O 35 may include a selector 56 and a NAND circuit 56a coupled to one input node of the selector 56.
  • the NAND circuit 56a may receive a HIT W signal and the signal from the selector 55.
  • the NAND circuit 56a may provide the signal from the selector 55 to the one input node of the selector 56 responsive to the active HIT W signal.
  • the selector 56 may have another input node that may receive the output signal of the buffer circuit 57 provided as the COREDATA.
  • the selector 56 may further receive the TEST from the mBIST 41 at a select node after inversion.
  • the selector 56 may provide the signals in parallel from the selector 55 via the buffer circuit 57, the same data as the COREDATA as BISTDATAin through the BISTDATA node to the DCMP 39.
  • the selector 56 may provide the EXDATA as the BISTDATAin through the BISTDATA node to the DCMP 39.
  • the TEST and the HIT W both are inactive. Thus, no data may be provided from the selector 45.
  • the EXDATA may be provided as the COREDATA through the selector 55 and the buffer circuit 57.
  • a selector 54 may receive the BISTDATAout and the COREDATA. The selector 54 may further receive the HIT R at a select node from the mBIST 41. The selector 55 may provide the BISTDATAout if the HIT_R is active. The selector 54 may further provide the COREDATA if the HIT R is inactive. A transmitter buffer 51 may provide the output signal in series as the EXDATA.
  • Fig. 6 is a schematic diagram of the access signal output circuit (ASO) 36 of Fig. 3, in accordance with an embodiment of the present disclosure.
  • the ASO 36 may include a receiver buffer Rx 58.
  • the receiver buffer Rx 58 receives the EXC/A from the AT 38.
  • the selector 60 may further receive the TC/A from the mBIST 41 at another input node.
  • the selector 60 may further receive the TEST from the mBIST 41 at a select node.
  • the TEST may be active in the test operation and inactive in the normal operation. If the TEST is inactive, the selector 60 may provide the EXC/A as the COREC/A.
  • the selector 60 may provide the TC/A as the COREC/A.
  • An AND circuit 61 may receive output signals of the selector 60.
  • the AND circuit 61 may further receive the TEST from the mBIST 41 after inversion.
  • the TEST after the inversion become inactive responsive to the active TEST, and the BISTC/A node becomes inactive.
  • the AND circuit 61 may provide the EXC/A received from the selector 60 responsive to the inactive TEST to the BISTC/A.
  • Figs. 7A and 7B are schematic diagrams of a portion of the storage area MEM 412 of Fig. 3, in accordance with an embodiment of the present disclosure.
  • the MEM 412 may include error catch memories (ECMs) 70.
  • ECMs 70 may receive test address information TA included in the MEMCTL and the fail signal P/F.
  • the ECMs 70 may store the TA as failure information including defective address information, responsive to the active fail signal P/F.
  • the ECMs 70 may have a plurality of portions, including a portion 71a and a portion 71b.
  • the portion 71a of the plurality of portions of ECMs 70 may store the failure information.
  • the failure information may be detected and stored during the initialization operation, as earlier described.
  • the defective address information DEFAIF may be provided from the portion 71a to a corresponding ACMP 40.
  • the portion 71b of the plurality of portions of ECMs 70 may include spare memory cells that may replace defective memory cells of the core chips 22 addressed by the failure information.
  • the portion 71b of the plurality of portions of ECMs 70 may receive a control signal R/WCTL in the MEMCTL indicative of the read operation or the write operation.
  • the portion 71b may include one or more spare memory cells that may store data from the BISTDATA node responsive to the TA and the R/WCTL indicative of the write operation, in the write operation.
  • the portion 71b may provide data from the one or more spare memory cells to the BISTDATA node responsive to the TA and the R/WCLTL indicative of the read operation, in the read operation.
  • Fig. 7C is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
  • the ECMs 70 may further include a page buffer PB 72, which may buffer data between the spare memory cells in the portion 71b and the DI/O 35.
  • the PB 72 may include a plurality of flip-flops, for example, which may provide access speed higher than access speed of the ECMs 70.
  • Fig. 7D is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
  • the ECMs 70 may further include anti-fuses AF 73 to permanently store the failure information.
  • Fig. 8 is a block diagram of the I/F chip 31 including a memory Built-in Self Test
  • the HITD 49' may control read and write operations on the MEM 412' instead of the mBISTL 411 '.
  • the HITD 49' may provide a portion of the MEMCTL signals, such as the TA and the R/WCTL to the MEM 412', when all ACMR[m:0] are indicative of a match between the access address information (e.g., read address information) and the DEFAIF[m:0].
  • Fig. 9 is a schematic diagram of one core chip of the plurality of core chips 22 in the semiconductor device in accordance with an embodiment of the present disclosure.
  • the one core chip may include an access control circuit AC 80, a memory cell array MA 82, a spare memory cell array SMA 83 and a defective address storing circuit AF 81.
  • the defective address storing circuit AF 81 may include and fuses.
  • memory cells in the MA 82 and memory cells in the SMA 83 may be different in type from the memory cells in the MEM 412 of the I/F chip 21.
  • the memory cells in the MA 82 and the memory cells in the SMA 83 may be DRAM memory cells.
  • the SMA may include a plurality of spare memory cells to replace the defective memory cells in the memory cell array as described earlier.
  • the AF 81 may store failure information, including defective address information.
  • the MEM 412 of the I/F chip 21 may also store defective address information. The defective address information stored in the MEM 412 of the I/F chip 21 and the defective address information stored in the AF 81 of each core chip 22 may be different from each other.
  • the defective address information detected during a test operation may be stored into the AF 81 of each core chip 22 while the defective address information detected during an initialization operation may be stored into the MEM 412 of the I/F chip 21.
  • the AF 81 may store the defective address information for memory cells in the MA 82 on the same core chip 22, while the MEM 412 of the I/F chip 21 may store the defective address information for cells placed in any core chip 22 in the semiconductor device.
  • the AC 80 may perform read and write operations on the MA 82 in response to the COREC/A via the TSVA 34 and the COREDATA via the TSVD 33.
  • the AC 80 may access memory cells in SMA 83 when the COREC/A includes the access address information that corresponds with one or more addresses included of the defective address information in DEFADDC provided from the AF 81.
  • Fig. 10 is a block diagram of the I/F chip including a memory Built-in Self Test (mBIST) circuit in the semiconductor device in accordance with an embodiment of the present disclosure.
  • the I/F chip 21 may include a command decoder 1002, which may be included in the access signal output circuit ASO 36 in Fig. 3.
  • the command decoder 1002 may receive command/address signals CA via an input buffer Rx, and may further provide write command information (WriteCom), read command information (ReadCom) and an address to one or more core chips 22 through a signal line 23 implemented as a through substrate via (TSV) (e.g., TSVA 34 in Fig. IB).
  • TSV through substrate via
  • the I/F chip 21 may include a controller circuit 1001 and an mBIST circuit 4F which includes an mBIST logic circuit 41 F and a storage area MEM 412'.
  • the MEM 412' may include a plurality of memory circuits 1004 and 1005.
  • the plurality of memory circuits 1004 and 1005 may be static random access memories (SRAMs) and one memory circuit 1004 of the plurality of memory circuits 1004 and 1005 may include a CAM memory that may function in a contents- addressable-memory (CAM) mode.
  • the controller circuit 1001 may control the memory circuit 1004 through a flag memory 1003.
  • the controller circuit 1001 may receive the WritcCom and the ReadCom.
  • the controller circuit 1001 may also receive HIT signal from a storage area MEM 412'.
  • the controller circuit 1001 may also receive flag information from a flag memory 1003.
  • the flag information may include in-use flag information and locked flag information.
  • the in-use flag information may indicate that an area with the CAM mode corresponding to a particular address provided in the memory circuit 1004 is already in use.
  • the locked flag information may indicate that an area in the CAM mode corresponding to a particular address provided in the memory circuit 1004 is locked and unmodifiable (e.g., already storing the defective address information).
  • the memory circuit 1005 may provide data stored on the memory circuit 1005 on a RAMDAT node (e.g., the BISTDATA node in Fig. 3).
  • a multiplexer MUX 1006 and a multiplexer MUX 1007 which may function as the DI/O[n:0] 35 in Fig. 3, receive the data from the RAMDAT node.
  • the MUX 1007 may receive data read from the core chip 22 via a TSV 23 (e.g., TSVD 33 in Fig. IB) via a ReadData node.
  • the MUX 1007 provides an output signal to a data queue DQ node via an output buffer Tx, responsive to the HIT signal from the memory circuit 1004. If the HIT signal is active (e.g., a logic high level) indicative that the RAMDAT is to be provided, the MUX 1007 may provide the data from the RAMDAT node.
  • the MUX 1007 may provide the read out data from the core chip 22 via the ReadData node if the HIT signal is inactive (e.g., a logic low level) and no replacement data is stored in the MEM 412' for the particular address.
  • the MUX 1006 receives the data from the RAMDAT node as well as test data from the mBIST logic circuit 41 F and provides either the data from the RAMDAT node or the test data as expected data on an EXPDAT (e.g., the EXP in Fig.
  • a comparator 1008 responsive to whether the data is to be matched with the data from the MEM 412' or the test data from the mBIST logic circuit 41 F (e.g., the test write data TWDATA in Fig. 3).
  • the comparator 1008 e.g., the DCMP 39 in Fig. 3 compares the expected data the EXPDAT node and the ReadData and provides comparison result signal CMPRSLT (e.g., the P/F in Fig. 3) to the controller circuit 1001 and the mBIST logic circuit 41 ⁇ .
  • Either the controller circuit 1001 or the mBIST logic circuit 411 ' may send control signals (Ctrl) to set the in-use flag information and the locked flag information in the flag memory 1003.
  • the controller circuit 1001 may write an address provided along with the WriteCom into the memory circuit 1004 in response to the WriteCom, the in-use flag information, the locked flag information and the HIT signal.
  • the memory circuit 1004 may store the address in the CAM memory and may further provide an inactive HIT signal to the memory circuit 1005 and the controller circuit 1001, if the address is not stored in the CAM memory in the memory circuit 1004.
  • the memory circuit 1004 may provide an active HIT signal to the memory circuit 1005 and the controller circuit 1001, if the address is already stored in the CAM memory in the memory circuit 1004.
  • the memory circuit 1005 may store data provided from the data queue DQ node via an input buffer Rx on a WriteData node, responsive to the HIT signal.
  • the memory circuit 1005 may store data on the WriteData node in an area addressed by a newly allocated address, if the HIT signal is inactive.
  • the memory circuit 1005 may store the data on the WriteData node in an area already allocated corresponding to the address, if the HIT signal is active.
  • Data on the WriteData node may be also provided to respective memory cells of the core chip 22. In some embodiments, data on the WriteData node may not be provided to the respective memory cells of the core chip 22 if the corresponding address have been held and locked in the CAM memory.
  • the controller circuit 1001 may prevent writing the address if the in-use flag information or the locked flag information for all addresses in the memory circuit 1004 may indicate that all of the CAM memory in the memory circuit 1004 is already in-use or locked and thus unavailable.
  • Fig. 11A is a simplified flow diagram of a write operation in the I/F chip of Fig. 10, in accordance with an embodiment of the present disclosure.
  • the controller circuit 1001 may provide the address along with the WriteCom to the flag memory 1003 and the flag memory 1003 may provide the address to the memory circuit 1004 on a CAMWE signal.
  • the memory circuit 1004 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001 and the memory circuit 1005.
  • the memory circuit 1005 may check whether the HIT signal is active (SI 101) and may further store the data on the WriteData node if the HIT signal is active (SI 106).
  • the controller circuit 1001 may check the in-use flag information and the locked flag information (SI 102). If the in-use flag information and the locked flag information are indicative that all the CAM memory is used (CAM overflow), the controller circuit 1001 may skip writing the address in the CAM memory in the memory circuit 1004 and may end the writing operation (SI 107).
  • the controller circuit 1001 may skip writing the address in the CAM memory in the memory circuit if the address is related to system defined conditions (SI 103).
  • the system defined conditions may include an address sampling.
  • the address sampling condition may include random sampling, the addresses which tend to be accessed frequently, the addresses in a same area (e.g., the same core, the same channel, etc.) having a frequent repair history and thus a higher deficiency rate than other area, or the addresses in a predetermined range due to system configuration (e.g., having a small margin).
  • an additional counter having a count of the additional counter may be changed at a predetermined interval to point a next address in the core dies to be stored for checking the deficiency.
  • it is possible to write the address to overwrite an oldest address stored in the CAM memory (after skipping the step SI 102).
  • the controller circuit 1001 may write the address to the CAM memory (SI 105) and set in-use flag information of the address.
  • the memory circuit 1005 stores the data on the WriteData node along with the address and the WriteCom (SI 106) if the HIT signal is active or if the controller circuit 1001 writes the address to the CAM memory and the write operation is completed (SI 107).
  • the memory circuit 1004 may compare the address provided along with the ReadCom with addresses stored in the CAM memory.
  • the memory circuit 1004 may provide the active HIT signal to the memory circuit 100S if the address is found in the addresses in the CAM memory.
  • the multiplexer MUX 1007 may provide either the data on the RAMDAT node or ReadData from the core chip 22 via the TSV 23 to the data queue DQ via an output buffer Tx, responsive to the active HIT signal.
  • the comparator 1008 may compare the ReadData with the data on the RAMDAT node and may further provide the CMPRSLT signal.
  • the controller 1001 may control the flag memory 1003 to change the in-use flag information of the address inactive (e.g., "0"), if the CMPRSLT signal indicates that the data on the RAMDAT node match the ReadData.
  • the CAM memory which has stored the address is freed and becomes available to store a new address.
  • locked flag information of the address active e.g., "1" to prevent the CAM memory of the address from overwriting a new address.
  • the CAM memory may store the address as the defect address information and the read/write operation of the address in the core chip 22 may be redirected permanently to the memory circuit 100S in the MEM 412'.
  • Fig. 1 IB is a simplified flow diagram of a read operation in the l/F chip of Fig. 10, in accordance with an embodiment of the present disclosure.
  • the controller circuit 1001 may provide the address along with the ReadCom to the flag memory 1003 and the flag memory 1003 may provide the address to the memory circuit 1004 on a CAM WE signal.
  • the memory circuit 1004 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001, the memory circuit 1005 and the multiplexer MUX 1007.
  • the MUX 1007 may check whether the HIT signal is active (SI 111) and may further provide the ReadData from the core chip 22 via the TSV 23, if the HIT signal is inactive "N" (SI 112) and the write operation is completed (SI 118). If the HIT signal is active "Y", the MUX 1007 may provide the data on the RAMDAT node to the data queue DQ (SI 113). The comparator 1008 may compare the ReadData with the data on the RAMDAT node with and may further provide the CMPRSLT signal (SI 116).
  • the controller 1001 may control the flag memory 1003 to change the in-use flag information of the address inactive (e.g., "0"), if the CMPRSLT signal indicates that the data on the RAMDAT node match the ReadData (SI 116) and the read operation is completed (SI 118). Thus, the CAM memory which has stored the address is freed and becomes available to store a new address. If the CMPRSLT signal indicates that the data on the RAMDAT node do not match the ReadData, the flag memory 1003 may set locked flag information of the address active (e.g., " 1") to prevent the CAM memory of the address from overwriting a new address (S 1117) and the read operation is completed (Si l l 8).
  • the flag memory 1003 may set locked flag information of the address active (e.g., " 1") to prevent the CAM memory of the address from overwriting a new address (S 1117) and the read operation is completed (Si l l 8).
  • Fig. 12 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure. Description of components corresponding to components included in and previously described with reference to Fig. 10 will not be repeated.
  • the I/F chip 21 may include a storage area MEM 1200 including a plurality of memory circuits 1204 and 1205.
  • the plurality of memory circuits 1204 and 1205 may be static random access memories (SRAMs) and one memory circuit 1204 of the plurality of memory circuits 1204 and 1205 may include a CAM memory that may function in a contents-addressable-memory (CAM) mode.
  • the controller circuit 1001 may control the memory circuit 1204 through a flag memory 1003.
  • the controller circuit 1001 may receive a WriteCom and a ReadCom.
  • the controller circuit 1001 may also receive HIT signal from a storage area MEM 1200.
  • the controller circuit 1001 may also receive flag information from a flag memory 1003.
  • the flag information may include in-use flag information and locked flag information.
  • the in-use flag information may indicate that an area with the CAM mode corresponding to a particular address provided in the memory circuit 1204 is already in use.
  • the locked flag information may indicate that an area in the CAM mode corresponding to a particular address provided in the memory circuit 1204 is locked and unmodifiable (e.g., already storing the defective address information).
  • the memory circuit 1205 may provide data stored on the memory circuit 1205 on a RAMDAT node (e.g., the BISTDATA node in Fig. 3).
  • a multiplexer MUX 1006 and a multiplexer MUX 1007 which may function as the DI/O[n:0] 35 in Fig. 3, receive the data from the RAMDAT node.
  • the MUX 1007 may receive data read from the core chip 22 via a TSV 23 (e.g., TSVD 33 in Fig. IB) via a ReadData node.
  • the MUX 1007 provides an output signal to a data queue DQ node via an output buffer Tx, responsive to the HIT signal from the memory circuit 1204.
  • the MUX 1007 may provide the data from the RAMDAT node.
  • the MUX 1007 may provide the read out data from the core chip 22 via the ReadData node if the HIT signal is inactive (e.g., a logic low level) and no replacement data is stored in the memory circuit 1205 for the particular address.
  • the MUX 1006 receives the data from the RAMDAT node as well as test data from the memory circuit 1205 and provides either the data from the RAMDAT node or the test data as expected data on an EXPDAT (e.g., the EXP in Fig.
  • the comparator 1008 (e.g., the DCMP 39 in Fig. 3) compares the expected data the EXPDAT node and data read from the core chip 22 on the ReadData node and provides comparison result signal CMPRSLT (e.g., the P/F in Fig. 3) to the controller circuit 1001.
  • the controller circuit 1001 may send control signals (Ctrl) to set the in-use flag information and the locked flag information in the flag memory 1003.
  • An I/F chip may include an error correction code (ECC) function.
  • Fig. 13 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure. Description of components corresponding to components included in and previously described with reference to Fig. 10 will not be repeated.
  • the I/F chip 21 may include a storage area MEM 1300 including a plurality of memory circuits 1304 and 1305.
  • the plurality of memory circuits 1304 and 1305 may be static random access memories (SRAMs) and one memory circuit 1304 of the plurality of memory circuits 1304 and 130S may include a CAM memory that may function in a contents-addressable-memory (CAM) mode.
  • SRAMs static random access memories
  • CAM contents-addressable-memory
  • the memory circuit 1304 may include two ports Portl 1321 and Port2 1322.
  • the port Portl 1321 may receive an address from a command decoder 1002 for accessing the CAM memory.
  • the port Port2 1322 may receive a control signal CAM WE and an error address signal ERRADD.
  • the memory circuit 130S may include two ports Portl 1323 and Port2 1324.
  • the port Portl 1323 may receive a WritcCom from the command decoder 1002 and may further receive data on a WriteData node from a data queue DQ node in a write operation.
  • the port Portl 132 may receive a ReadCom from the command decoder 1002.
  • the port Port2 1322 may receive the control signal CAMWE and a read data signal REDDAT.
  • the controller circuit 1001 may provide a control signal Ctrl to control the memory circuit 1304 through a flag memory 1003.
  • the flag memory 1003 may provide the control signal CAMWE to the port Port2 1322 of the memory circuit 1304 and the plurality of memory circuit 1304 and 1305.
  • the controller circuit 1001 may receive HIT signal from the storage area MEM 1300.
  • the controller circuit 1001 may also receive flag information from the flag memory 1003.
  • the flag information may include in-use flag information and locked flag information.
  • the in-use flag information may indicate mat an area with the CAM mode corresponding to a particular address provided in the memory circuit 1304 is already in use.
  • the locked flag information may indicate that an area in the CAM mode corresponding to a particular address provided in the memory circuit 1304 is locked and unmodifiable (e.g., already storing the defective address information).
  • the memory circuit 1305 may provide data stored on the memory circuit 1305 on a RAMDAT node (e.g., the BISTDATA node in Fig. 3) through the port Portl 1323.
  • the I/F die 21 may include a write error correction circuit (WECC) 1310 which may generate a Write error correction code (WriteECC) based on data on the WriteData node during a write operation and provide the ECC on a WriteECC node to the core chip 22 with the data on the WriteData node.
  • WECC write error correction circuit
  • the I/F die 21 may include read error correction circuit (RECC) 1311 that may receive the data on the ReadData node and a Read error correction code (ReadECC) from the core chip 22 via a TSV 23 (e.g., TSVD 33 in Fig.
  • the RECC 1311 circuit may activate an ERR signal if there is the error, regardless of whether error is correctable or uncorrectable.
  • the MUX 1306 may receive data REDDAT which is a data signal from the RECC 1311 and the data on the RAMDAT node.
  • the MUX 1306 provides an output signal to the data queue DQ node via an output buffer Tx, responsive to the HIT signal from the memory circuit 1304. If the HIT signal is active (e.g., a logic high level) indicative that data on the RAMDAT node is to be provided, the MUX 1306 may provide the data from the RAMDAT node.
  • the MUX 1306 may provide the read out data REDDAT which may be after the error correction at the RECC 1311, if the HIT signal is inactive (e.g., a logic low level) and no replacement data is stored in the memory circuit 1305 for the particular address.
  • the I/F die 21 may also include a first-in-first-out memory (FIFO) 1309 which may receive the address and store the address during read operations responsive to the ReadCom.
  • the FIFO 1309 provides the address on an ERRADD node to the port Port2 1322 of the memory circuit 1304 responsive to the ERR signal.
  • Fig. 14A is a simplified flow diagram of a write operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure.
  • the controller circuit 1001 may provide the address along with the WriteCom to the flag memory 1003 and the flag memory 1003 may provide the address to the plurality of the memory circuits 1304 on the CAMWE signal.
  • the memory circuit 1304 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001, the memory circuit 130S.
  • the memory circuit 130S may check whether the HIT signal is active (SI 401) and may further store the data on the WriteData node if the HIT signal is active (SI 402) and end the writing operation (SI 403). If the HIT signal is not active "N", the controller circuit 1001 may skip writing the data and end the writing operation (S 1403).
  • Fig. 14B is a simplified flow diagram of a read operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure.
  • the controller circuit 1001 may provide the address along with the ReadCom to the flag memory 1003 and the flag memory 1003 may provide the address to the memory circuit 1304 on the CAMWE signal.
  • the memory circuit 1304 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001, the memory circuit 1005 and the multiplexer MUX 1306.
  • the MUX 1306 may check whether the HIT signal is active (S1411).
  • the MUX 1306 may provide the data on the RAMDAT node to the data queue DQ (S1412). and the read operation is completed (S1417).
  • the MUX 1306 may provide the data on the REDDAT node from the RECC 1311 to the data queue DQ (S1413), if the HIT signal is inactive "N”.
  • the controller circuit 1001 may check whether the ERR signal from the RECC 1311 is active (S1414). If the ERR signal from the RECC 1311 is not active "N”, the read operation is completed (S 1417). If the ERR signal is active "Y”, the controller circuit 1001 may check the in-usc flag information and the locked flag information, responsive to the ERR signal (SI 415).
  • the controller circuit 1001 may skip writing the address in the CAM memory in the memory circuit 1304 and may end the writing operation (S1417). If the CAM memory is available (CAM overflow - ' ⁇ "), the controller circuit 1001 may control the plurality of memory circuits 1304 and 130S by providing the CAMWE signal. The memory circuit 1304 may write the address to the CAM memory from the FIFO 1309 via the ERRADD node
  • the memory circuit 1305 may store the data on the REDDAT node (S1416).
  • the controller circuit 1001 may further set in-usc flag information of the address (SI 416). After the step in S 1416, the read operation is completed

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • For Increasing The Reliability Of Semiconductor Memories (AREA)
  • Tests Of Electronic Circuits (AREA)

Abstract

Apparatuses and methods of sharing error correction memory on an interface chip are described. An example apparatus includes: at least one memory chip having a plurality of first memory cells and an interface chip coupled to the at least one memory chip and having a control circuit and a storage area. The control circuit detects one or more defective memory cells of the first memory cells of the at least one memory chip. The control circuit further stores first defective address information of the one or more defective memory cells of the first memory cells into the storage area. The interface chip responds to the first defective address information and an access request to access the storage area in place of the at least one memory chip when the access request has been provided with respect to the one or more defective memory cells of the first memory cells.

Description

SHARED ERROR DETECTION AND CORRECTION MEMORY
BACKGROUND
[001] High data reliability, high speed of memory access, lower power consumption and reduced chip size are features that are demanded from semiconductor memory. In recent years, three-dimensional (3D) memory devices have been introduced. Some 3D memory devices are formed by stacking chips (e.g., dice) vertically and interconnecting the chips using through substrate vias (TSVs). Benefits of the 3D memory devices include shorter interconnects which reduce circuit delays and power consumption, a large number of vertical vias between layers which allow wide bandwidth buses between functional blocks in different layers, and a considerably smaller footprint Thus, the 3D memory devices contribute to higher memory access speed, lower power consumption and chip size reduction. Example 3D memory devices include Hybrid Memory Cube (HMC), High Bandwidth Memory (HBM), and a wide-I/O dynamic random access memory (DRAM).
[002] For example, High Bandwidth Memory (HBM) is a type of memory including a high- performance DRAM interface and vertically stacked DRAM. A typical HBM stack of four DRAM chips (e.g., core chips) has two 128-bit channels per chip for a total of eight input/output channels and a width of 1024 bits in total. An interface (I/F) chip of the HBM provides an interface with the eight input/output channels, which function independently of each other. For example, a clock frequency, a command sequence, and data can be independently provided for each channel. Thus, the eight input/output channels are not necessarily synchronous to each other.
[003] There are several types of tests which may be performed for the HBM. For example, a type of test can be performed using a memory Built-in Self Test (mBIST) circuit that may be provided on the I/F chip. The mBIST circuit is provided for verifying failures resulting from stacking the chip. The mBIST circuit may include a memory to store defect information called an error-catch memory (ECM). Using the defect information, for example, hard repair such as blowing fuse to disconnect rows and columns with faulty bits and replacing them with redundant rows or columns may be performed.
[004] The HBM has a post package repair function performed by using the mBIST circuit The post package repair function uses redundancy cells for repair and these redundancy cells are normally formed in memory matrices of the core. However, the post package repair function may not be able to repair defects if the number of defective cells is greater than a number of repairable cells by providing redundancy cells. Furthermore, it may be difficult to repair one or more defective cells which are redundancy cells.
SUMMARY
[005] An example apparatus according to an embodiment of the disclosure may include at least one memory chip and an interface chip. The at least one memory chip may include a plurality of first memory cells. Then interface chip may be coupled to the at least one memory chip and may include a control circuit and a storage area. The control circuit may detect one or more defective memory cells of the first memory cells of the at least one memory chip and may further store first defective address information of the one or more defective memory cells of the first memory cells into the storage area. The interface chip may respond, at least in part, to the first defective address information and an access request to access the storage area in place of the at least one memory chip when the access request has been provided with respect to the one or more defective memory cells of the first memory cells.
[006] An example interface chip apparatus according to an embodiment of the disclosure may include a test circuit including a test logic circuit and one or more memories. The test logic circuit may detect one or more defective memory cells of at least one memory chip when the test circuit is coupled to the at least one memory chip. A first portion of the one or more memories may store first address information of the one or more defective memory cells as defective address information. The test circuit may access one or more memory cells in a second portion of the one or more memories responsive to an access request, when the test logic circuit determines that the access request includes second address information relevant to the one or more defective memory cells based on the defective address information.
[007] An example method of sharing error correction memory on an interface chip, according to an embodiment of the disclosure may include detecting one or more defective memory cells of a memory chip; storing first address information of the one or more defective memory cells as defective address information; and accessing one or more memory cells in in the error correction memory responsive to an access request including second address information relevant to the one or more defective memory cells based on the defective address information. BRIEF DESCRIPTION OF THE DRAWINGS
[008] Fig. 1A is a schematic diagram of an interface (I/F) chip and a plurality of core chips in a semiconductor device in accordance with an embodiment of the present disclosure.
[009] Fig. IB is a schematic diagram of a memory system including the semiconductor device that includes the interface (I/F) chip and the plurality of core chips in accordance with an embodiment of the present disclosure.
[010] Fig. 2 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure.
[011] Fig. 3 is a block diagram of the interface circuit in Fig. 2 including a memory Built-in
Self Test (mBIST) circuit in the semiconductor device in accordance with an embodiment of the present disclosure.
[012] Fig. 4A is a simplified flow diagram of a test operation process by the mBIST in Fig.
3 in the semiconductor device in accordance with an embodiment of the present disclosure.
[013] Fig. 4B is a simplified flow diagram of a test operation as a part of an initialization operation process performed by the mBIST in the semiconductor device in accordance with an embodiment of the present disclosure.
[014] Fig. 5 is a schematic diagram of the data input/output circuit of Fig. 3, in accordance with an embodiment of the present disclosure.
[015] Fig. 6 is a schematic diagram of the access signal output circuit of Fig. 3, in accordance with an embodiment of the present disclosure.
[016] Fig. 7A is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
[017] Fig. 7B is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
[018] Fig. 7C is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
[019] Fig. 7D is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure.
[020] Fig. 8 is a block diagram of the I/F chip including a memory Built-in Self Test
(mBIST) circuit in the semiconductor device in accordance with an embodiment of the present disclosure.
[021] Fig. 9 is a schematic diagram of one core chip of the plurality of core chips in the semiconductor device in accordance with an embodiment of the present disclosure. [022] Fig. 10 is a block diagram of the I/F chip including a memory Built-in Self Test (mBIST) circuit in the semiconductor device in accordance with an embodiment of the present disclosure.
|023] Fig. 1 1A is a simplified flow diagram of a write operation in the I/F chip of Fig. 10, in accordance with an embodiment of the present disclosure.
[024] Fig. 1 IB is a simplified flow diagram of a read operation in the I/F chip of Fig. 10, in accordance with an embodiment of the present disclosure.
[025] Fig. 12 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure.
[026] Fig. 13 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure.
[027] Fig. 14A is a simplified flow diagram of a write operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure.
[028] Fig. 14B is a simplified flow diagram of a read operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
[029] Various embodiments of the present invention will be explained below in detail with reference to the accompanying drawings. The following detailed description refers to the accompanying drawings that show, by way of illustration, specific aspects and embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. Other embodiments may be utilized, and structure, logical and electrical changes may be made without departing from the scope of the present invention. The various embodiments disclosed herein are not necessary mutually exclusive, as some disclosed embodiments can be combined with one or more other disclosed embodiments to form new embodiments.
[030] Fig. 1A is a schematic diagram of an interface (I/F) chip and a plurality of core chips in a semiconductor device in accordance with an embodiment of the present disclosure. For example, the semiconductor device 10 may be a 3D memory device, such as an HBM, an HMC, a Wide-IO DRAM etc. The semiconductor device is formed by stacking chips vertically, as shown in Fig. 1A. The stacked chips may include an interface chip 21 and core chips 22. In this example, each core chip 22 may be a memory chip which includes two channels. Each channel may include a plurality of memory cells and circuitries accessing the memory cells. For example, the memory cells may be DRAM memory cells.
[031] Fig. IB is a schematic diagram of a memory system including the semiconductor device that includes the interface (I/F) chip and the plurality of core chips in accordance with an embodiment of the present disclosure. The memory system 1 may include a memory controller 11 and the 3D memory device 10. In this example, the core chips 22a, 22b, 22c and 22d include channels A and C, channels B and D, channels E and G, and channels F and H, respectively. As shown in Fig. IB, the channels A, B, C, D, E, F, G and H of the core chips 22 may be coupled to the I/F chip 21 via different signal lines 23a, 23 b, 23c, 23d, 23e, 23f, 23g, and 23h, respectively. The I/F chip 21 may include test control terminals 24, data terminals 25 and access signal terminals 26 which receive corresponding signals from a memory controller 11. For example, the test control terminals 24 may receive test control signals. The data terminals 25 may receive write data from the memory controller 11 or may transmit read data to the memory controller 1 1. The access signal terminals 26 may receive access signals from the memory controller 11. The access signals may include an operation command (e.g., a read command, a write command) and address information corresponding to the operation command.
[032] Fig. 2 is a block diagram of an interface circuit on the I/F chip 21 in the semiconductor device in accordance with an embodiment of the present disclosure. An interface circuit 31 is provided on the I/F chip 21. The interface circuit 31 may include channel interface blocks CIF_A 32a, CEF_B 32b, ... and CIF_H 32h which are coupled to corresponding channels, channel A, channel B and channel H of the core chip 22 in Fig.
IB. Furthermore, the interface circuit 31 may include a plurality of data through substrate vias, TSVDs 33 and access through substrate vias TSVAs 34. The TSVDs 33 and TSVAs 34 are configured to couple the channel interface blocks CIF_A 32a, CIFJB 32b, ... and CIF_H
32h to the channels, channel A, channel B and channel H of the core chip 22 in Fig. IB, respectively. In one embodiment, each signal line 23 in Fig. IB may include the corresponding TSVD 33 and TSVA 34 for each channel. Each channel interface block of the channel interface blocks CIF_A 32a, CIF_B 32b, ... and CIF_H 32h may include a data input/output circuit DI/O 35 and an access signal output circuit ASO 36. Each DI/O 35 may receive data from the coupled channel via the corresponding TSVD 33. The DI/O 35 may be further coupled to data terminals DT 37. The data terminals DT 37 may be the data terminals 25 in Fig. IB. The ASO 36 may be coupled to access signal terminals AT 38. The access signal terminals AT 38 may receive access signals including access requests provided externally (e.g., from the memory controller 11 of Fig. IB). Each access request may include a command including an operation requested (e.g., a read request, a write request, etc.) and an address where the operation requested by the command is operated. The access signal terminals AT 38 may be the access signal terminals 26 in Fig. IB. Each ASO 36 may provide the access requests as command signals and address signals, for example, to the coupled channel via the corresponding TSVA 34. In a normal read operation, a read request and corresponding address information are externally provided to the ASO 36 via the AT 38. The ASO 36 may provide the read request and the corresponding address information to the coupled channel via the TSVA 34, responsive to the read request The coupled channel may provide the data to the DI/O 35 via the TSVD 33 in response to the read request and the corresponding address information. The DI/O 35 may provide the data via the DT 37. Similarly, in a normal write operation, a write request and corresponding address information are externally provided to the ASO 36 via the AT 38. The ASO 36 may provide the write request and the corresponding address information to the coupled channel via the TSVA 34, responsive to the write request. At the same time, the DI/O 35 may receive write data from the DT 37 and provide the write data to the coupled channel via the TSVD 33 responsive to the write request and the corresponding address information. Each of the channel interface blocks CIF_A 32a, CIF_B 32b, ... and CIF_H 32h may further include a data comparator circuit DCMP 39 and an access signal comparator circuit ACMP 40. Functionalities of the DCMP 39 and the ACMP 40 will be provided later in detail.
|033] The interface circuit 31 may further include a memory built-in self-test block mBIST 41. The mBIST 41 may provide test signals to the channel interface blocks CIF_A 32a, CEF B 32b, ... and CIF H 32h via a multiplexer MUX 45 in order to perform test operations on each channel responsive to test control signals TCTL. The TCTL are externally provided via test control terminals TCT 43. The TCT 43 may be the test control terminals 24 in Fig. IB. The mBIST 41 may perform defective cell detection and repair functions during an initialization operation and a normal operation following the initialization operation, such as read and write operations, as well as during a test operation. Functionalities of the mBIST 41 will be provided later in detail. The interlace circuit 31 may further include a power detection circuit PD 42. The PD 42 may provide a power on signal PON to the mBIST 41 responsive to power supply PS externally provided via a power terminal PT 44. The mBIST 41 may start the initialization operation in response to the PON. The interface circuit 31 may further include test terminals TT 46 and test pads TP 47. The TT 46 may be coupled to a socket or the like in order to couple an external tester (not shown) to the interface circuit 31. The TP 47 may couple an external tester (not shown) having a probe card interface, for example, to the interface circuit 31. The TP 47 may be used for a test during wafer process. The MUX 45 may select one of the TT 46, the TP 47 and the mBIST 41 as a source of the test signals and provide the test signals to the channel interface blocks CDF_A 32a, CIFJB 32b, ... and CIF_H 32h.
[034] In Fig. 2, the terminals, such as the DT 37, the AT 38, the TCT 43, the PT 44 and the
TT 46 represented by circles may be formed by micro bump electrodes, for example. The test pad TP 47 may be formed by a pad electrode, for example. Each of the DT 37, the AT 38, the TCT 43, the PT 44 and the TT 46 may include one or more terminals. The TP 47 may include one or more pads.
[035] Fig. 3 is a block diagram of the interface circuit 31 in Fig. 2 including the memory Built-in Self Test (mBIST) circuit 41 in the semiconductor device in accordance with an embodiment of the present disclosure. Fig. 3 shows a connection between mBIST 41 of Fig. 2 and one of the channel interface blocks CIF_A 32a, CIF_B 32b, ... and CIF_H 32h of Fig. 2 as an example. The other channel interface blocks may be coupled to the mBIST 41 similarly as shown in Fig. 3. The interface chip may include a plurality of TSVD[n:0] 33 and corresponding DI/O[n:0] 35, DT[n:0] 37 and DCMP[n:0] 39provided with a single channel. Similarly, the interface chip may include a plurality of TSVA[m:0] 34, ASO[m:0] 36, AT[m:0] 38 and ACMP[m:0] 40provided with a single channel. In Fig. 3, "m" is a number of bits for command/address signals and "n" is a number of bits for data signals, and Mm" and "n" may be different from each other. The mBIST 41 may include an mBIST logic circuit mBISTL 411 and a storage area MEM 412. The mBISTL 411 may be a control circuit, such as an algorithmic pattern generator (APG), which may control operations of the mBIST 41. The MEM 412 may be a single memory circuit. Alternatively, the MEM 412 may be a plurality of memory circuits, each of which may perform independent functions individually. For example, the MEM 412 may include error catch memories (ECMs) and microcode memories (MCMs). For example, each of the ECMs and the MCMs may include static random access memories (SRAMs). The MCMs may store microcodes. The microcodes may represent test patterns for testing the memory cells of the core chips. The mBISTL 411 may perform test operations in accordance with the microcodes. The ECMs may store defective address information during the test operations. For example, the MEM 412, such as the ECMs and/or the MCMs, can be used as spare cells for repairing defective memory cells in the core chips 22 as described later in detail. In some embodiments, the I/F chip 21 may further include a read only memory (ROM) to store the microcodes. In another embodiment, initial states of the MCMs in the MEM 412 may represent the microcodes.
[036] In one embodiment, the mBIST 41 may operate during the test operations. The mBIST 41 may further operate during the initialization operation and the normal operation. In particular, the mBIST 41 may detect one or more defective cells among memory cells of core chips 22. Upon detection of the one or more defective cells, the mBIST 41 may store address information for the defective cells in one portion of the MEM 412, for example the ECMs and/or MCMs, during the initialization operation. Furthermore, in the normal operation, the mBIST 41 may use an other portion of the MEM 412, for example, the ECMs and/or MCMs, to replace the defective cells with the other portion of the MEM 412 which may serve as spare memory. In other words, the mBIST 41 may redirect access to the defective cells of the core chips 22 to the other portion of the MEM 412.
[037] In some embodiments, each of the core chips 22 may further include a defective address storing circuit, such as anti-fuses and spare memory cells. When the core chips 22 include the defective address storing circuit, the other portion of the MEM 412 of the I/F chip 21 and the spare memory cells of the core chips 22 may be used for repairing different defective cells from each other. In some embodiments, the other portion of the MEM 412 of the I/F chip 21 may replace the defective cells of the spare memory cells of the core chips 22. The detailed operation of the mBIST 41 and the channel interface block 32 during each of the test operation, the initialization operation and the normal operation including a read operation or a write operation are described below.
1) Test operation
[038] Fig. 4A is a simplified flow diagram of a test operation process by the mBIST in Fig.
3 in the semiconductor device in accordance with an embodiment of the present disclosure. For example, the test operation process may be performed by the mBIST before shipping.
[039] In an operation block 401, the mBIST 41 may receive the test control signals TCTL externally provided through the test control terminals TCT 43. The test control signals may include a test instruction. In response to the TCTL, the mBISTL 411 may execute the microcodes to perform a test operation. In some embodiments, the microcodes may be prestored on the MCMs or a ROM of the I/F chip 21 as their initial states. In other embodiments, the MCMs may store the microcodes externally provided. The mBISTL 411 may provide a test enable signal TEST to the DI/O[n:0] 35 and the ASO[n:0] 36. The DI/O[n:0] 35 may couple, in response to the TEST, the TSVD[n:0] 33 to a BISTDATA node coupled to the DCMP[n:0] 39 and the mBIST 41 while decoupling the external data terminal DT[n:0] 37 from the DCMP[n:0] 39 and the mBIST 41. Similarly, the ASO[m:0] 36 may couple, in response to the TEST, the TSVA[m:0] 34 to a BIST command and address node BISTC/A coupled to the ACMP[m:0] 40 and the mBIST 41 while decoupling the external access terminal AT[m:0] 38 from the ACMP[m:0] 40 and the mBIST 41. The detailed description of the DI/O[n:0] 35 and the ASO[m:0] 36 will be provided later referring to Figs. 5 and 6.
[040] In an operation block 402, the mBISTL 411 may provide test write data
TWDATA[n:0] to the DI/O[n:0] 35. The mBISTL 411 may provide a test command/address signal TC/A[m:0] including test write commands and test address information to the ASO[m:0] 36. Accordingly, the DI/O[n:0] 35 and the ASO[m:0] 36 may perform test write operations on the respective channel. For example, the DI/O[n:0] may provide the TWDATA[n:0J as core data COREDATA[n:0] to the TSVD[n:0] 33. The ASO[m:0] 36 may provide the TC/A[m:0] as a core command and address signal COREC/A[m:0] to the TSVA[m:0] 34.
[041] After the COREDATA[n:0] are stored in memory cells designated by the COREC/A[m:0] in the respective channel of the core chips 22, the mBISTL 411 may provide the test command/address signal TC/A[m:0] test including read commands and test address information to the ASO[m:0] 36 to perform test read operations in an operation block 403. The mBISTL 411 may provide a memory control signal MEMCTL including the test address information TA to the MEM 412. Accordingly, the ASO[m:0] 36 may provide the COREC/Aim:01 including the test read commands and the test address information to the respective channel via the TSVA[m:0] 34. The respective channel may provide the COREDATA[n:0] including test read data to the TSVD[n:0] 33 responsive to the COREC/A[m:0]. The DI/O[n:0] 35 may receive the test read data and provide the test read data to the DCMP[n:0] 39.
[042] In an operation block 404, the DCMP[n:0] 39 may compare the test read data with the expected data EXP[n:0] that is the corresponding test write data provided from the mBISTL 411. The DCMP[n:0] 39 may provide a comparison result DCMR[n:0] to an error detection circuit PFD 48. If at least one DCMR[n:0] is indicative of data discrepancy between the test read data and the EXP[n:0], the PFD 48 may provide an active fail detection signal P/F (e.g., at a logic high level) in an operation block 406. If there is no data discrepancy between the test read data and the EXP[n:0] for all the DCMR[n:0], the PFD 48 may provide an inactive P/F (e.g., at a logic low level). In response to the active P/F, the MEM 412 may store the corresponding test address information as defective address information in an operation block 406.
[043] After testing the memory cells of the respective channel, determined defective address information stored in the MEM 412 may be programed into a defective address storing circuit, for example anti-fuses, of the core chips 22 in an operation block 407. The determined defective address information may be read out from the MEM 412 to an external tester or an external controller (not shown) before programming in an operation block 40S. The programing of the defective address information may be performed by mBISTL 411 or externally by the tester or the controller.
2i Initialization operation
|044] Fig. 4B is a simplified flow diagram of a test operation as a part of an initialization operation process performed by the mBIST in the semiconductor device in accordance with an embodiment of the present disclosure. The test operation as the part of the initialization operation process is similar to the test operation before shipping described above except a power on sequence. In an operation block 421, the mBISTL 411 may respond to a power on signal PON instead of the TCTL in order to initiate testing of the memory cells. In some embodiments, the TCTL may be used to initiate testing in an operational block 40 F that is similar to the operation block 401 in Fig. 4A. In the test operation as the part of the initialization operation process, the programming of the determined defective address information into the defective address storing circuit of the core chips 22 may be skipped. Instead, the determined defective address information may be stored in the MEM 412 of the I/F chip 21 in an operation block 406', after executing the test write in an operation block 402', similar to the operation block 402, the test read in an operation block 403' similar to the operation block 403 and the comparison of the test read data and the expected data EXP[n:0] in an operation block 404' that is similar to the operation block 404. In an operation block 422, the normal access operation, including a write operation and a read operation, may follow the initialization operation.
3¾ Normal write operation
[045] In the normal operation, the TEST may be inactive (e.g., at a logic low level). The DI/O[n:0] 35 may couple the TSVD[n:0] 33 to the DT[n:0] 37 until hit signals Hit_R/W become active (e.g., at a logic high level). The detailed description of the DI/O[n:0] 35 will be provided later referring to Fig. 5. An external memory controller (e.g., the memory controller 11 of Fig. IB) may provide external data signals EXDATA[n:0] including write data to the DT[n:0] 37. The external memory controller may also provide an EXC/A[m:0] signal including a write command and access address information (e.g., write address information) to the AT[m:0] 38. The ASO[m:0] 36 may provide the EXC/A[m:0] as the COREC/A[m:0] to the respective channel. In this embodiment, the ASO[m:0] 36 may further provide the EXC/A[m:0] on the BISTC/A node to the ACMP[m:0] 40. The ACMP[m:0] 40 may compare the access address information on the BISTC/A node with the defective address information DEFAIF[m:0] provided from the MEM 412. The ACMP[m:0] 40 may provide a comparison result ACMRfmiO] to a hit detection circuit HITD 49. When all ACMR[m:0] are indicative of a match between the access address information and the DEFAIF[m:0], the HITD 49 may provide a HIT signal to the mBISTL 411. In some embodiments, the MEM 412 may provide the DEFAIF[m:0] to the ACMP[m:0] 40. The ACMP[m:0] 40 may include latch circuits, for example, which may hold the DEFAIF[m:0] at least during the normal operation (e.g., the write operation or the read operation). The HIT signal may indicate that the provided address information corresponds to the defective address information. The HIT signal may also indicate a type of the operation (e.g., the write operation). In response to the HIT, the mBIST 41 may provide a HIT W signal to the DI/O[n:0] 35. Responsive to the HIT_W signal, the DI/O[n:0] 35 provides the EXDATA[n:0] on the BISTDATA node to the MEM 412. The mBISTL 411 may perform the write operation on the MEM 412 to store the write date into the MEM. In some embodiments, the mBISTL 411 may stop performing a write operation on the respective channel on the core chip 22 while the repair operation, such as redirecting the access to the MEM 412, is being performed. When the address information on the BISTC/A node does not correspond to the DEFAIF[m:0], the DI/O[n:0] 35 may provide the EXDATA[n:0] to the channel as the COREDATA[n:0] and the COREDATA[n:0] are written into the corresponding memory cells of the core chip 22.
4) Normal read operation
[046] The normal read operation is similar to the normal write operation described above except a reading sequence. In the read operation, the mBISTL 411 may provide the HIT R signal in response to the HIT signal provided from the HITD 49 when any ACMR[m:0] is indicative of a match between the access address information (e.g., read address information) and the DEFAJLF[m:0]. Responsive to the HITJR. signal, the DI/O[n:0] 35 may couple the DT[n:0] 37 to the MEM 412 through die BISTDATA node. The mBISTL may further perform the read operation on the MEM 412 to provide requested read data from the MEM 412. In some embodiments, the mBISTL 41 1 may control a read latency for the read data. The read data retrieved from the MEM 412 may be provided to the DI/O[n:0] 35 via the BISTDATA node. The DI/O[n:0] 35 may selectively couple the BISTDATA node to the EXDATA node responsive to the active HIT_R signal, thus the DI/O[n:0] 35 may provide the read data to the DT[n:0] 37. In some embodiments, the mBISTL 411 may stop performing a read operation on the respective channel on the core chip 22 while the repair operation, such as redirecting the access to the MEM 412 for reading, is being performed. When the address information on the BISTC/A node does not correspond to the DEFAJF[m:0], the data may be retrieved from the memory cells of the respective channel of core chip 22 to the DT[n:0] 37.
[047] Fig. 5 is a schematic diagram of the data input/output circuit (DI/O) 35 of Fig. 3, in accordance with an embodiment of the present disclosure. The DI/O 35 is coupled to the EXDATA node. The DI/O 35 may include a receiver buffer Rx 50. The receiver buffer Rx 50 receives the EXDATA from the DT 37 and may further provide the EXDATA to one input node of a selector 55. The selector 55 may further receive BISTDATAout on the BISTDATA node from the DCMP 39 at another input node. The selector 55 may further receive the TEST from the mBIST 41 at a select node after inversion. The TEST may be active in the test operation and inactive in the normal operation (e.g., a write operation, a read operation, etc.). If the TEST is inactive, the selector 55 may provide the EXDATA to a buffer circuit 57. If the TEST is active, the selector 55 may provide the BISTDATAout from the BISTDATA node to the buffer circuit 57. For example, the buffer circuit 57 may be a tri- state buffer. An input node of the buffer circuit 57 may receive a signal from the selector 55. An enable input node of the buffer circuit 57 may receive a write enable signal CORE W, indicative of enabling writing data to the core chips 22. For example, the CORE_W may be provided as a portion of the COREC/A by the ASO 36 in response to a write command. The buffer circuit 57 may provide the signal from the selector 55 as the COREDATA when the CORE_W is active (e.g., a logic high level). The DI/O 35 may include a selector 56 and a NAND circuit 56a coupled to one input node of the selector 56. The NAND circuit 56a may receive a HIT W signal and the signal from the selector 55. The NAND circuit 56a may provide the signal from the selector 55 to the one input node of the selector 56 responsive to the active HIT W signal. The selector 56 may have another input node that may receive the output signal of the buffer circuit 57 provided as the COREDATA. The selector 56 may further receive the TEST from the mBIST 41 at a select node after inversion. When the TEST is active, the selector 56 may provide the signals in parallel from the selector 55 via the buffer circuit 57, the same data as the COREDATA as BISTDATAin through the BISTDATA node to the DCMP 39. When the HIT_W is active while the TEST is inactive, the selector 56 may provide the EXDATA as the BISTDATAin through the BISTDATA node to the DCMP 39. In the normal write operation without defective address information, the TEST and the HIT W both are inactive. Thus, no data may be provided from the selector 45. Thus, merely the EXDATA may be provided as the COREDATA through the selector 55 and the buffer circuit 57.
[048] A selector 54 may receive the BISTDATAout and the COREDATA. The selector 54 may further receive the HIT R at a select node from the mBIST 41. The selector 55 may provide the BISTDATAout if the HIT_R is active. The selector 54 may further provide the COREDATA if the HIT R is inactive. A transmitter buffer 51 may provide the output signal in series as the EXDATA.
[049] Fig. 6 is a schematic diagram of the access signal output circuit (ASO) 36 of Fig. 3, in accordance with an embodiment of the present disclosure. The ASO 36 may include a receiver buffer Rx 58. The receiver buffer Rx 58 receives the EXC/A from the AT 38. The selector 60 may further receive the TC/A from the mBIST 41 at another input node. The selector 60 may further receive the TEST from the mBIST 41 at a select node. The TEST may be active in the test operation and inactive in the normal operation. If the TEST is inactive, the selector 60 may provide the EXC/A as the COREC/A. If the TEST is active, the selector 60 may provide the TC/A as the COREC/A. An AND circuit 61 may receive output signals of the selector 60. The AND circuit 61 may further receive the TEST from the mBIST 41 after inversion. The TEST after the inversion become inactive responsive to the active TEST, and the BISTC/A node becomes inactive. The AND circuit 61 may provide the EXC/A received from the selector 60 responsive to the inactive TEST to the BISTC/A.
[050] Figs. 7A and 7B are schematic diagrams of a portion of the storage area MEM 412 of Fig. 3, in accordance with an embodiment of the present disclosure. For example, the MEM 412 may include error catch memories (ECMs) 70. In the test operation, the ECMs 70 may receive test address information TA included in the MEMCTL and the fail signal P/F. The ECMs 70 may store the TA as failure information including defective address information, responsive to the active fail signal P/F. In the normal operation, the ECMs 70 may have a plurality of portions, including a portion 71a and a portion 71b. The portion 71a of the plurality of portions of ECMs 70 may store the failure information. The failure information may be detected and stored during the initialization operation, as earlier described. The defective address information DEFAIF may be provided from the portion 71a to a corresponding ACMP 40. The portion 71b of the plurality of portions of ECMs 70 may include spare memory cells that may replace defective memory cells of the core chips 22 addressed by the failure information. The portion 71b of the plurality of portions of ECMs 70 may receive a control signal R/WCTL in the MEMCTL indicative of the read operation or the write operation. The portion 71b may include one or more spare memory cells that may store data from the BISTDATA node responsive to the TA and the R/WCTL indicative of the write operation, in the write operation. The portion 71b may provide data from the one or more spare memory cells to the BISTDATA node responsive to the TA and the R/WCLTL indicative of the read operation, in the read operation.
|051] Fig. 7C is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure. In this embodiment, the ECMs 70 may further include a page buffer PB 72, which may buffer data between the spare memory cells in the portion 71b and the DI/O 35. The PB 72 may include a plurality of flip-flops, for example, which may provide access speed higher than access speed of the ECMs 70.
[052] Fig. 7D is a schematic diagram of a portion of the MEM of Fig. 3, in accordance with an embodiment of the present disclosure. In this embodiment, the ECMs 70 may further include anti-fuses AF 73 to permanently store the failure information.
|053] Fig. 8 is a block diagram of the I/F chip 31 including a memory Built-in Self Test
(mBIST) circuit 41 in the semiconductor device in accordance with an embodiment of the present disclosure. Description of components corresponding to components included in and previously described with reference to Fig. 3 will not be repeated. Unlike the I/F chip of Fig. 3, the HITD 49' may control read and write operations on the MEM 412' instead of the mBISTL 411 '. The HITD 49' may provide a portion of the MEMCTL signals, such as the TA and the R/WCTL to the MEM 412', when all ACMR[m:0] are indicative of a match between the access address information (e.g., read address information) and the DEFAIF[m:0]. In this example, the HITD 49' may directly provide the HIT_R and the HIT W to the DI/O[n:0] 35' instead of providing the HIT to the mBISTL 41 P and having the mBISTL 41 P providing the HIT R and the HIT W. [054] Fig. 9 is a schematic diagram of one core chip of the plurality of core chips 22 in the semiconductor device in accordance with an embodiment of the present disclosure. The one core chip may include an access control circuit AC 80, a memory cell array MA 82, a spare memory cell array SMA 83 and a defective address storing circuit AF 81. For example, the defective address storing circuit AF 81 may include and fuses. In some embodiments, memory cells in the MA 82 and memory cells in the SMA 83 may be different in type from the memory cells in the MEM 412 of the I/F chip 21. For example, the memory cells in the MA 82 and the memory cells in the SMA 83 may be DRAM memory cells. The SMA may include a plurality of spare memory cells to replace the defective memory cells in the memory cell array as described earlier. The AF 81 may store failure information, including defective address information. As described above, the MEM 412 of the I/F chip 21 may also store defective address information. The defective address information stored in the MEM 412 of the I/F chip 21 and the defective address information stored in the AF 81 of each core chip 22 may be different from each other. In some embodiments, the defective address information detected during a test operation may be stored into the AF 81 of each core chip 22 while the defective address information detected during an initialization operation may be stored into the MEM 412 of the I/F chip 21. In some embodiments, the AF 81 may store the defective address information for memory cells in the MA 82 on the same core chip 22, while the MEM 412 of the I/F chip 21 may store the defective address information for cells placed in any core chip 22 in the semiconductor device. The AC 80 may perform read and write operations on the MA 82 in response to the COREC/A via the TSVA 34 and the COREDATA via the TSVD 33. The AC 80 may access memory cells in SMA 83 when the COREC/A includes the access address information that corresponds with one or more addresses included of the defective address information in DEFADDC provided from the AF 81.
[055] Fig. 10 is a block diagram of the I/F chip including a memory Built-in Self Test (mBIST) circuit in the semiconductor device in accordance with an embodiment of the present disclosure. The I/F chip 21 may include a command decoder 1002, which may be included in the access signal output circuit ASO 36 in Fig. 3. The command decoder 1002 may receive command/address signals CA via an input buffer Rx, and may further provide write command information (WriteCom), read command information (ReadCom) and an address to one or more core chips 22 through a signal line 23 implemented as a through substrate via (TSV) (e.g., TSVA 34 in Fig. IB). The I/F chip 21 may include a controller circuit 1001 and an mBIST circuit 4F which includes an mBIST logic circuit 41 F and a storage area MEM 412'. The MEM 412' may include a plurality of memory circuits 1004 and 1005. For example, the plurality of memory circuits 1004 and 1005 may be static random access memories (SRAMs) and one memory circuit 1004 of the plurality of memory circuits 1004 and 1005 may include a CAM memory that may function in a contents- addressable-memory (CAM) mode. The controller circuit 1001 may control the memory circuit 1004 through a flag memory 1003. The controller circuit 1001 may receive the WritcCom and the ReadCom. The controller circuit 1001 may also receive HIT signal from a storage area MEM 412'. The controller circuit 1001 may also receive flag information from a flag memory 1003. For example, the flag information may include in-use flag information and locked flag information. The in-use flag information may indicate that an area with the CAM mode corresponding to a particular address provided in the memory circuit 1004 is already in use. The locked flag information may indicate that an area in the CAM mode corresponding to a particular address provided in the memory circuit 1004 is locked and unmodifiable (e.g., already storing the defective address information). The memory circuit 1005 may provide data stored on the memory circuit 1005 on a RAMDAT node (e.g., the BISTDATA node in Fig. 3). A multiplexer MUX 1006 and a multiplexer MUX 1007, which may function as the DI/O[n:0] 35 in Fig. 3, receive the data from the RAMDAT node. The MUX 1007 may receive data read from the core chip 22 via a TSV 23 (e.g., TSVD 33 in Fig. IB) via a ReadData node. The MUX 1007 provides an output signal to a data queue DQ node via an output buffer Tx, responsive to the HIT signal from the memory circuit 1004. If the HIT signal is active (e.g., a logic high level) indicative that the RAMDAT is to be provided, the MUX 1007 may provide the data from the RAMDAT node. The MUX 1007 may provide the read out data from the core chip 22 via the ReadData node if the HIT signal is inactive (e.g., a logic low level) and no replacement data is stored in the MEM 412' for the particular address. The MUX 1006 receives the data from the RAMDAT node as well as test data from the mBIST logic circuit 41 F and provides either the data from the RAMDAT node or the test data as expected data on an EXPDAT (e.g., the EXP in Fig. 3) node to a comparator 1008, responsive to whether the data is to be matched with the data from the MEM 412' or the test data from the mBIST logic circuit 41 F (e.g., the test write data TWDATA in Fig. 3). The comparator 1008 (e.g., the DCMP 39 in Fig. 3) compares the expected data the EXPDAT node and the ReadData and provides comparison result signal CMPRSLT (e.g., the P/F in Fig. 3) to the controller circuit 1001 and the mBIST logic circuit 41 Γ. Either the controller circuit 1001 or the mBIST logic circuit 411 ' may send control signals (Ctrl) to set the in-use flag information and the locked flag information in the flag memory 1003.
[056] In write operations, the controller circuit 1001 may write an address provided along with the WriteCom into the memory circuit 1004 in response to the WriteCom, the in-use flag information, the locked flag information and the HIT signal. For example, the memory circuit 1004 may store the address in the CAM memory and may further provide an inactive HIT signal to the memory circuit 1005 and the controller circuit 1001, if the address is not stored in the CAM memory in the memory circuit 1004. The memory circuit 1004 may provide an active HIT signal to the memory circuit 1005 and the controller circuit 1001, if the address is already stored in the CAM memory in the memory circuit 1004. The memory circuit 1005 may store data provided from the data queue DQ node via an input buffer Rx on a WriteData node, responsive to the HIT signal. For example, the memory circuit 1005 may store data on the WriteData node in an area addressed by a newly allocated address, if the HIT signal is inactive. The memory circuit 1005 may store the data on the WriteData node in an area already allocated corresponding to the address, if the HIT signal is active. Data on the WriteData node may be also provided to respective memory cells of the core chip 22. In some embodiments, data on the WriteData node may not be provided to the respective memory cells of the core chip 22 if the corresponding address have been held and locked in the CAM memory. The controller circuit 1001 may prevent writing the address if the in-use flag information or the locked flag information for all addresses in the memory circuit 1004 may indicate that all of the CAM memory in the memory circuit 1004 is already in-use or locked and thus unavailable.
1057] Fig. 11A is a simplified flow diagram of a write operation in the I/F chip of Fig. 10, in accordance with an embodiment of the present disclosure. Upon receiving a write command at the command decoder 1002 (SHOO), the controller circuit 1001 may provide the address along with the WriteCom to the flag memory 1003 and the flag memory 1003 may provide the address to the memory circuit 1004 on a CAMWE signal. The memory circuit 1004 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001 and the memory circuit 1005. The memory circuit 1005 may check whether the HIT signal is active (SI 101) and may further store the data on the WriteData node if the HIT signal is active (SI 106). If the HIT signal is not active "N", the controller circuit 1001 may check the in-use flag information and the locked flag information (SI 102). If the in-use flag information and the locked flag information are indicative that all the CAM memory is used (CAM overflow), the controller circuit 1001 may skip writing the address in the CAM memory in the memory circuit 1004 and may end the writing operation (SI 107).
[058] Optionally, the controller circuit 1001 may skip writing the address in the CAM memory in the memory circuit if the address is related to system defined conditions (SI 103). For example, the system defined conditions may include an address sampling. The address sampling condition may include random sampling, the addresses which tend to be accessed frequently, the addresses in a same area (e.g., the same core, the same channel, etc.) having a frequent repair history and thus a higher deficiency rate than other area, or the addresses in a predetermined range due to system configuration (e.g., having a small margin). Alternatively, it is possible to have an additional counter having a count of the additional counter may be changed at a predetermined interval to point a next address in the core dies to be stored for checking the deficiency. Alternatively, it is possible to write the address to overwrite an oldest address stored in the CAM memory (after skipping the step SI 102). Depending on the steps SI 102 and SI 103, the controller circuit 1001 may write the address to the CAM memory (SI 105) and set in-use flag information of the address. The memory circuit 1005 stores the data on the WriteData node along with the address and the WriteCom (SI 106) if the HIT signal is active or if the controller circuit 1001 writes the address to the CAM memory and the write operation is completed (SI 107).
[059] In read operations, the memory circuit 1004 may compare the address provided along with the ReadCom with addresses stored in the CAM memory. The memory circuit 1004 may provide the active HIT signal to the memory circuit 100S if the address is found in the addresses in the CAM memory. The multiplexer MUX 1007 may provide either the data on the RAMDAT node or ReadData from the core chip 22 via the TSV 23 to the data queue DQ via an output buffer Tx, responsive to the active HIT signal. As described earlier, the comparator 1008 may compare the ReadData with the data on the RAMDAT node and may further provide the CMPRSLT signal. The controller 1001 may control the flag memory 1003 to change the in-use flag information of the address inactive (e.g., "0"), if the CMPRSLT signal indicates that the data on the RAMDAT node match the ReadData. Thus, the CAM memory which has stored the address is freed and becomes available to store a new address. If the CMPRSLT signal indicates that the data on the RAMDAT node do not match the ReadData, locked flag information of the address active (e.g., "1") to prevent the CAM memory of the address from overwriting a new address. As a result, the CAM memory may store the address as the defect address information and the read/write operation of the address in the core chip 22 may be redirected permanently to the memory circuit 100S in the MEM 412'.
[060] Fig. 1 IB is a simplified flow diagram of a read operation in the l/F chip of Fig. 10, in accordance with an embodiment of the present disclosure. Upon receiving a read command at the command decoder 1002 (SHOO), the controller circuit 1001 may provide the address along with the ReadCom to the flag memory 1003 and the flag memory 1003 may provide the address to the memory circuit 1004 on a CAM WE signal. The memory circuit 1004 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001, the memory circuit 1005 and the multiplexer MUX 1007. The MUX 1007 may check whether the HIT signal is active (SI 111) and may further provide the ReadData from the core chip 22 via the TSV 23, if the HIT signal is inactive "N" (SI 112) and the write operation is completed (SI 118). If the HIT signal is active "Y", the MUX 1007 may provide the data on the RAMDAT node to the data queue DQ (SI 113). The comparator 1008 may compare the ReadData with the data on the RAMDAT node with and may further provide the CMPRSLT signal (SI 116). The controller 1001 may control the flag memory 1003 to change the in-use flag information of the address inactive (e.g., "0"), if the CMPRSLT signal indicates that the data on the RAMDAT node match the ReadData (SI 116) and the read operation is completed (SI 118). Thus, the CAM memory which has stored the address is freed and becomes available to store a new address. If the CMPRSLT signal indicates that the data on the RAMDAT node do not match the ReadData, the flag memory 1003 may set locked flag information of the address active (e.g., " 1") to prevent the CAM memory of the address from overwriting a new address (S 1117) and the read operation is completed (Si l l 8).
[061] Fig. 12 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure. Description of components corresponding to components included in and previously described with reference to Fig. 10 will not be repeated. The I/F chip 21 may include a storage area MEM 1200 including a plurality of memory circuits 1204 and 1205. For example, the plurality of memory circuits 1204 and 1205 may be static random access memories (SRAMs) and one memory circuit 1204 of the plurality of memory circuits 1204 and 1205 may include a CAM memory that may function in a contents-addressable-memory (CAM) mode. The controller circuit 1001 may control the memory circuit 1204 through a flag memory 1003. The controller circuit 1001 may receive a WriteCom and a ReadCom. The controller circuit 1001 may also receive HIT signal from a storage area MEM 1200. The controller circuit 1001 may also receive flag information from a flag memory 1003. For example, the flag information may include in-use flag information and locked flag information. The in-use flag information may indicate that an area with the CAM mode corresponding to a particular address provided in the memory circuit 1204 is already in use. The locked flag information may indicate that an area in the CAM mode corresponding to a particular address provided in the memory circuit 1204 is locked and unmodifiable (e.g., already storing the defective address information). The memory circuit 1205 may provide data stored on the memory circuit 1205 on a RAMDAT node (e.g., the BISTDATA node in Fig. 3). A multiplexer MUX 1006 and a multiplexer MUX 1007, which may function as the DI/O[n:0] 35 in Fig. 3, receive the data from the RAMDAT node. The MUX 1007 may receive data read from the core chip 22 via a TSV 23 (e.g., TSVD 33 in Fig. IB) via a ReadData node. The MUX 1007 provides an output signal to a data queue DQ node via an output buffer Tx, responsive to the HIT signal from the memory circuit 1204. If the HIT signal is active (e.g., a logic high level) indicative that data on the RAMDAT node is to be provided, the MUX 1007 may provide the data from the RAMDAT node. The MUX 1007 may provide the read out data from the core chip 22 via the ReadData node if the HIT signal is inactive (e.g., a logic low level) and no replacement data is stored in the memory circuit 1205 for the particular address. The MUX 1006 receives the data from the RAMDAT node as well as test data from the memory circuit 1205 and provides either the data from the RAMDAT node or the test data as expected data on an EXPDAT (e.g., the EXP in Fig. 3) node to a comparator 1008, responsive to whether the data is to be matched with the test data from the memory circuit 1205. The comparator 1008 (e.g., the DCMP 39 in Fig. 3) compares the expected data the EXPDAT node and data read from the core chip 22 on the ReadData node and provides comparison result signal CMPRSLT (e.g., the P/F in Fig. 3) to the controller circuit 1001. The controller circuit 1001 may send control signals (Ctrl) to set the in-use flag information and the locked flag information in the flag memory 1003.
|062] An I/F chip may include an error correction code (ECC) function. Fig. 13 is a block diagram of the I/F chip in the semiconductor device in accordance with an embodiment of the present disclosure. Description of components corresponding to components included in and previously described with reference to Fig. 10 will not be repeated. The I/F chip 21 may include a storage area MEM 1300 including a plurality of memory circuits 1304 and 1305. For example, the plurality of memory circuits 1304 and 1305 may be static random access memories (SRAMs) and one memory circuit 1304 of the plurality of memory circuits 1304 and 130S may include a CAM memory that may function in a contents-addressable-memory (CAM) mode. The memory circuit 1304 may include two ports Portl 1321 and Port2 1322. The port Portl 1321 may receive an address from a command decoder 1002 for accessing the CAM memory. The port Port2 1322 may receive a control signal CAM WE and an error address signal ERRADD. The memory circuit 130S may include two ports Portl 1323 and Port2 1324. The port Portl 1323 may receive a WritcCom from the command decoder 1002 and may further receive data on a WriteData node from a data queue DQ node in a write operation. The port Portl 132 may receive a ReadCom from the command decoder 1002. The port Port2 1322 may receive the control signal CAMWE and a read data signal REDDAT. The controller circuit 1001 may provide a control signal Ctrl to control the memory circuit 1304 through a flag memory 1003. The flag memory 1003 may provide the control signal CAMWE to the port Port2 1322 of the memory circuit 1304 and the plurality of memory circuit 1304 and 1305. The controller circuit 1001 may receive HIT signal from the storage area MEM 1300. The controller circuit 1001 may also receive flag information from the flag memory 1003. For example, the flag information may include in-use flag information and locked flag information. The in-use flag information may indicate mat an area with the CAM mode corresponding to a particular address provided in the memory circuit 1304 is already in use. The locked flag information may indicate that an area in the CAM mode corresponding to a particular address provided in the memory circuit 1304 is locked and unmodifiable (e.g., already storing the defective address information). The memory circuit 1305 may provide data stored on the memory circuit 1305 on a RAMDAT node (e.g., the BISTDATA node in Fig. 3) through the port Portl 1323. A multiplexer MUX 1306, which may function as the DI/O[n:0] 35 in Fig. 3, receive the data from the RAMDAT node.
[063] The I/F die 21 may include a write error correction circuit (WECC) 1310 which may generate a Write error correction code (WriteECC) based on data on the WriteData node during a write operation and provide the ECC on a WriteECC node to the core chip 22 with the data on the WriteData node. The I/F die 21 may include read error correction circuit (RECC) 1311 that may receive the data on the ReadData node and a Read error correction code (ReadECC) from the core chip 22 via a TSV 23 (e.g., TSVD 33 in Fig. IB), check if there is an error based on the ReadECC and the data read from the core chip 22 on the ReadData node, and may further correct the error, if any, during read operations. The RECC 1311 circuit may activate an ERR signal if there is the error, regardless of whether error is correctable or uncorrectable.
1064] The MUX 1306 may receive data REDDAT which is a data signal from the RECC 1311 and the data on the RAMDAT node. The MUX 1306 provides an output signal to the data queue DQ node via an output buffer Tx, responsive to the HIT signal from the memory circuit 1304. If the HIT signal is active (e.g., a logic high level) indicative that data on the RAMDAT node is to be provided, the MUX 1306 may provide the data from the RAMDAT node. The MUX 1306 may provide the read out data REDDAT which may be after the error correction at the RECC 1311, if the HIT signal is inactive (e.g., a logic low level) and no replacement data is stored in the memory circuit 1305 for the particular address. The I/F die 21 may also include a first-in-first-out memory (FIFO) 1309 which may receive the address and store the address during read operations responsive to the ReadCom. The FIFO 1309 provides the address on an ERRADD node to the port Port2 1322 of the memory circuit 1304 responsive to the ERR signal.
[065] Fig. 14A is a simplified flow diagram of a write operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure. Upon receiving a write command at the command decoder 1002 (SI 400), the controller circuit 1001 may provide the address along with the WriteCom to the flag memory 1003 and the flag memory 1003 may provide the address to the plurality of the memory circuits 1304 on the CAMWE signal. The memory circuit 1304 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001, the memory circuit 130S. The memory circuit 130S may check whether the HIT signal is active (SI 401) and may further store the data on the WriteData node if the HIT signal is active (SI 402) and end the writing operation (SI 403). If the HIT signal is not active "N", the controller circuit 1001 may skip writing the data and end the writing operation (S 1403).
Fig. 14B is a simplified flow diagram of a read operation in the I/F chip of Fig. 13, in accordance with an embodiment of the present disclosure. Upon receiving a read command at the command decoder 1002 (S1410), the controller circuit 1001 may provide the address along with the ReadCom to the flag memory 1003 and the flag memory 1003 may provide the address to the memory circuit 1304 on the CAMWE signal. The memory circuit 1304 may determine whether the address is already stored in the CAM memory and may further provide the HIT signal to the controller circuit 1001, the memory circuit 1005 and the multiplexer MUX 1306. The MUX 1306 may check whether the HIT signal is active (S1411). If the HIT signal is active "Y", the MUX 1306 may provide the data on the RAMDAT node to the data queue DQ (S1412). and the read operation is completed (S1417). The MUX 1306 may provide the data on the REDDAT node from the RECC 1311 to the data queue DQ (S1413), if the HIT signal is inactive "N". The controller circuit 1001 may check whether the ERR signal from the RECC 1311 is active (S1414). If the ERR signal from the RECC 1311 is not active "N", the read operation is completed (S 1417). If the ERR signal is active "Y", the controller circuit 1001 may check the in-usc flag information and the locked flag information, responsive to the ERR signal (SI 415). If the in-use flag information and the locked flag information are indicative that all the CAM memory is used (CAM overflow="Y"), the controller circuit 1001 may skip writing the address in the CAM memory in the memory circuit 1304 and may end the writing operation (S1417). If the CAM memory is available (CAM overflow - 'Ν"), the controller circuit 1001 may control the plurality of memory circuits 1304 and 130S by providing the CAMWE signal. The memory circuit 1304 may write the address to the CAM memory from the FIFO 1309 via the ERRADD node
(51416) , responsive to the CAMWE signal. The memory circuit 1305 may store the data on the REDDAT node (S1416). The controller circuit 1001 may further set in-usc flag information of the address (SI 416). After the step in S 1416, the read operation is completed
(51417) . Thus, it is possible to provide remedy with the CAM memory based on the result of ECC function.
[066] Logic levels of signals used in the embodiments described the above are merely examples. However, in other embodiments, combinations of the logic levels of signals other than those specifically described in the present disclosure may be used without departing from the scope of the present disclosure.
[067] Although this invention has been disclosed in the context of certain preferred embodiments and examples, it will be understood by those skilled in the art that the inventions extend beyond the specifically disclosed embodiments to other alternative embodiments and/or uses of the inventions and obvious modifications and equivalents thereof. In addition, other modifications which are within the scope of this invention will be readily apparent to those of skill in the art based on this disclosure. It is also contemplated that various combination or sub-combination of the specific features and aspects of the embodiments may be made and still fall within the scope of the inventions. It should be understood that various features and aspects of the disclosed embodiments can be combined with or substituted for one another in order to form varying mode of the disclosed invention. Thus, it is intended that the scope of at least some of the present invention herein disclosed should not be limited by the particular disclosed embodiments described above.

Claims

CLAIMS What is claimed is:
1. An apparatus comprising:
at least one memory chip comprising a plurality of first memory cells; and
an interlace chip coupled to the at least one memory chip and comprising a control circuit and a storage area,
wherein the control circuit is configured to detect one or more defective memory cells of the first memory cells of the at least one memory chip and further configured to store first defective address information of the one or more defective memory cells of the first memory cells into the storage area, and
wherein the interface chip is configured to respond, at least in part, to the first defective address information and an access request to access the storage area in place of the at least one memory chip when the access request has been provided with respect to the one or more defective memory cells of the first memory cells.
2. The apparatus as claimed in claim 1, wherein the control circuit is configured to perform a memory test operation in response to a power on signal of the apparatus, and further configured to detect the one or more defective memory cells of the first memory cells in the memory test operation.
3. The apparatus as claimed in claim 2, wherein the interface chip further comprises a terminal that is configured to receive a test instruction,
wherein the control circuit is configured to perform the memory test operation in response to the test instruction.
4. The apparatus as claimed in claim 1, wherein the control circuit is further configured to perform a latency control on access the storage area.
5. The apparatus as claimed in claim 1, wherein the at least one memory chip comprises a defective address storing circuit in which second defective address information is stored, the second defective address information of an other one or more defective memory cells of the first memory cells of the at least one memory chip, wherein the first defective address information and the second defective address information are different from each other.
6. The apparatus as claimed in claim S, wherein the control circuit is configured to detect the other one or more defective memory cells of the first memory cells to produce the second defective address information configured to store the second defective address information in the storage area, and further configured to program the second defective address information into the defective address storing circuit of the at least one memory chip.
7. The apparatus as claimed in claim 1 , wherein the storage area comprises a plurality of second memory cells that are different in type from the first memory cells.
8. The apparatus as claimed in claim 1, wherein the at least one memory chip and the interface chip are stacked with each other.
9. The apparatus as claimed in claim 1, wherein the control circuit is configured to store access address information and corresponding data into the storage area along with providing the corresponding data to the at least one memory chip in a write operation and configured to respond to a read request to one or more cells of the first memory cells designated by the access address information to compare the corresponding data stored in the storage area and the corresponding data provided from the at least one memory chip to determine whether the one or more cells of the first memory cells designated by the access address information is defective or not.
10. The apparatus as claimed in claim 1, further comprising an error detection circuit configured to receive data from one or more cells of the first memory cells and provide an error signal when the data include an error, and wherein the control circuit is configured to respond to the error signal to identify the one or more cells as the one or more defective cells and store access address information designating the one or more cells into the storage area as the first defective address information.
11. The apparatus of claim 3, further comprising:
a memory controller configured to provide the access request and write data to the interface chip and further configured to receive read data from the interface chip, wherein the memory controller is further configured to provide the test instruction to the terminal of the interface chip.
12. An interface chip comprising:
a test circuit comprising:
a test logic circuit; and
one or more memories,
wherein the test logic circuit is configured to detect one or more defective memory cells of at least one memory chip when the test circuit is coupled to the at least one memory chip,
wherein a first portion of the one or more memories are configured to store first address information of the one or more defective memory cells as defective address information, and
wherein the test circuit is configured to access one or more memory cells in a second portion of the one or more memories responsive to an access request, when the test logic circuit determines that the access request includes second address information relevant to the one or more defective memory cells based on die defective address information.
13. The interface chip of claim 12, further comprising:
an access signal output circuit configured to provide an operation command and third address information to the at least one memory chip;
a data input/output circuit configured to receive test read data from the at least one memory chip and further configured to provide the test read data;
a data comparator circuit coupled to the data input/output circuit, configured to compare the test read data with expected data and further configured to provide a first comparison result; and
an error detection circuit coupled to the data comparator circuit and configured to provide a first detection signal responsive to the first comparison result to the test circuit, wherein the test logic circuit is configured to provide a test read command as the operation command and test address information as the address information to the access signal output circuit, and further configured to provide the expected data to the data comparator circuit, wherein the access signal output circuit is configured to provide die test read command and the test address information to the at least one memory chip, and
wherein the test circuit is configured to detect the one or more defective memory cells of the at least one memory chip based on the first comparison result.
14. The interface chip of claim 13, wherein the test logic circuit is further configured to provide test write data to the data input/output circuit and further configured to provide a test write command as the operation command and the test address information as the address information to the access signal output circuit prior to providing the test read command,
wherein the access signal output circuit configured to provide the test write command and the test address information to the at least one memory chip prior to providing the test read command,
wherein the data input/output circuit is further configured to provide the test write data to the at least one memory chip prior to receiving the test read data, and
wherein the expected data is the test write data.
15. The interface chip of claim 13, further comprising:
an access signal comparator circuit coupled to the access signal output circuit, configured to compare the test address information with the defective address information and further configured to provide a second comparison result; and
a hit detection circuit coupled to the access signal comparator circuit and configured to provide a second detection signal responsive to the second comparison result to the test circuit,
wherein the test logic circuit is configured to determine that the access request includes address information relevant to the one or more defective memory cells based on the defective address information responsive to the second detection signal.
16. The interface chip of claim 13, further comprising a test control terminal coupled to a memory controller,
wherein the test circuit is configured to receive configured to receive test control signals from the memory controller through the test control terminal, and further configured to configured to detect the one or more defective memory cells of the at least one memory chip responsive to the test control signals.
17. The interface chip of claim 13, further comprising a memory that is configured to store one or more microcodes,
wherein the test logic circuit is configured to execute the one or more microcodes to detect the one or more defective memory cells of the at least one memory chip.
18. A method of sharing error correction memory on an interface chip, comprising:
detecting one or more defective memory cells of a memory chip;
storing first address information of the one or more defective memory cells as defective address information; and
accessing one or more memory cells in in the error correction memory responsive to an access request including second address information relevant to the one or more defective memory cells based on the defective address information.
19. The method of claim 18, wherein detecting the one or more defective memory cells of the memory chip comprises:
providing test write data, test access signal including test write commands and test address information responsive to test control signals;
providing the test write data, test write commands and test address information to the memory chip including a memory cell corresponding to the test address information;
providing read commands and the test address information to a channel on the memory chip corresponding to the test address information;
receiving test read data from the channel;
comparing the test read data with expected data; and
providing a comparison result, and
wherein storing the first address information of the detected one or more defective memory cells as the defective address information is executed responsive to the comparison result.
20. The method of claim 19, further comprising:
receiving a power on signal,
wherein the defective address information is stored in a defective address storing circuit responsive to the power on signal.
21. The method of claim 20, wherein the defective address storing circuit is in the memory chip.
22. The method of claim 19, further comprising:
receiving the test control signals including a test instruction,
wherein the defective address information is stored in the error correction memory responsive to the test control signals.
PCT/US2017/035946 2016-06-15 2017-06-05 Shared error detection and correction memory WO2017218227A1 (en)

Priority Applications (6)

Application Number Priority Date Filing Date Title
KR1020207030188A KR102399014B1 (en) 2016-06-15 2017-06-05 Shared error detection and correction memory
CN202210673440.8A CN114974393A (en) 2016-06-15 2017-06-05 Interface chip equipment and method for sharing correction memory on interface chip
KR1020217030505A KR20210123403A (en) 2016-06-15 2017-06-05 Shared error detection and correction memory
EP17813792.3A EP3472708A4 (en) 2016-06-15 2017-06-05 Shared error detection and correction memory
CN201780037240.0A CN109313596B (en) 2016-06-15 2017-06-05 Interface chip equipment and method for sharing correction memory on interface chip
KR1020187036759A KR102170322B1 (en) 2016-06-15 2017-06-05 Shared error detection and correction memory

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US15/183,654 2016-06-15
US15/183,654 US10395748B2 (en) 2016-06-15 2016-06-15 Shared error detection and correction memory

Publications (1)

Publication Number Publication Date
WO2017218227A1 true WO2017218227A1 (en) 2017-12-21

Family

ID=60659779

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2017/035946 WO2017218227A1 (en) 2016-06-15 2017-06-05 Shared error detection and correction memory

Country Status (5)

Country Link
US (2) US10395748B2 (en)
EP (1) EP3472708A4 (en)
KR (3) KR20210123403A (en)
CN (2) CN109313596B (en)
WO (1) WO2017218227A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10395748B2 (en) 2016-06-15 2019-08-27 Micron Technology, Inc. Shared error detection and correction memory

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102378819B1 (en) * 2018-02-20 2022-03-25 삼성전자주식회사 Memory device
US11151006B2 (en) * 2018-07-02 2021-10-19 Samsung Electronics Co., Ltd. HBM RAS cache architecture
KR102587648B1 (en) * 2018-07-23 2023-10-11 삼성전자주식회사 Stacked memory devices, memory systems including the same and methods of testing the stacked memory devices
KR102703487B1 (en) * 2018-08-03 2024-09-06 에스케이하이닉스 주식회사 Data storage device and operating method thereof
WO2020152231A1 (en) * 2019-01-22 2020-07-30 Advantest Corporation Automated test equipment for testing one or more devices under test, method for automated testing of one or more devices under test, and computer program using a buffer memory
KR102657584B1 (en) 2019-05-20 2024-04-15 삼성전자주식회사 Wafer level test methods of semiconductor device using internal test enable signal
US10916489B1 (en) * 2019-10-02 2021-02-09 Micron Technology, Inc. Memory core chip having TSVS
US11705214B2 (en) * 2020-03-30 2023-07-18 Micron Technologv. Inc. Apparatuses and methods for self-test mode abort circuit
WO2022252205A1 (en) * 2021-06-04 2022-12-08 Yangtze Memory Technologies Co., Ltd. Firmware repair for three-dimensional nand memory
KR20240126736A (en) 2023-02-14 2024-08-21 국립부경대학교 산학협력단 Permanent defect qubit repair system and method using built-in-self-repair model at quantum circuit level

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6801471B2 (en) * 2002-02-19 2004-10-05 Infineon Technologies Ag Fuse concept and method of operation
US20080117696A1 (en) * 2006-11-20 2008-05-22 Chuan-Jen Chang Method for repairing defects in memory and related memory system
US20080141067A1 (en) * 2006-12-07 2008-06-12 Arm Limited Memory device and method of controlling access to such a memory device
US20140337573A1 (en) * 2013-05-08 2014-11-13 Qualcomm Incorporated Redirecting data from a defective data entry in memory to a redundant data entry prior to data access, and related systems and methods
US8924903B2 (en) * 2010-11-30 2014-12-30 Ps4 Luxco S.A.R.L. Semiconductor device having plural memory chip

Family Cites Families (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW374951B (en) * 1997-04-30 1999-11-21 Toshiba Corp Semiconductor memory
JP2000030483A (en) * 1998-07-15 2000-01-28 Mitsubishi Electric Corp Bist circuit for large-scale memory
JP2001051957A (en) * 1999-08-04 2001-02-23 Hitachi Ltd On-chip multiplexer
JP4002378B2 (en) * 1999-12-27 2007-10-31 エルピーダメモリ株式会社 Electronic circuit
US7046561B1 (en) * 2003-04-16 2006-05-16 Michael Tooher Memory compiler redundancy
US7359261B1 (en) * 2005-05-23 2008-04-15 Marvell International Ltd. Memory repair system and method
US8069377B2 (en) * 2006-06-26 2011-11-29 Micron Technology, Inc. Integrated circuit having memory array including ECC and column redundancy and method of operating the same
US20080011769A1 (en) * 2006-07-11 2008-01-17 Tempo Tool And Manufacturing Company Limited Media distribution system
JP4353336B2 (en) * 2006-12-26 2009-10-28 エルピーダメモリ株式会社 Semiconductor memory device and program method thereof
KR100909902B1 (en) * 2007-04-27 2009-07-30 삼성전자주식회사 Flash memory device and Flash memory system
US20100162037A1 (en) * 2008-12-22 2010-06-24 International Business Machines Corporation Memory System having Spare Memory Devices Attached to a Local Interface Bus
JP2011081732A (en) * 2009-10-09 2011-04-21 Elpida Memory Inc Semiconductor device, adjusting method for the same, and data processing system
JP5649888B2 (en) * 2010-09-17 2015-01-07 ピーエスフォー ルクスコ エスエイアールエルＰＳ４ Ｌｕｘｃｏ Ｓ．ａ．ｒ．ｌ． Semiconductor device
US8971094B2 (en) * 2011-09-16 2015-03-03 Inphi Corporation Replacement of a faulty memory cell with a spare cell for a memory circuit
US8804394B2 (en) * 2012-01-11 2014-08-12 Rambus Inc. Stacked memory with redundancy
KR20130098039A (en) * 2012-02-27 2013-09-04 삼성전자주식회사 Semiconductor device capable of rescuing defective characteristics after packaging
US9158619B2 (en) 2012-03-30 2015-10-13 Intel Corporation On chip redundancy repair for memory devices
US20130336051A1 (en) * 2012-04-24 2013-12-19 Being Advanced Memory Corporation Multibit memory with read voltage qualification at startup
US9128822B2 (en) * 2012-06-22 2015-09-08 Winbond Electronics Corporation On-chip bad block management for NAND flash memory
US9640279B1 (en) * 2012-09-12 2017-05-02 Cadence Design Systems, Inc. Apparatus and method for built-in test and repair of 3D-IC memory
US9087614B2 (en) * 2012-11-27 2015-07-21 Samsung Electronics Co., Ltd. Memory modules and memory systems
AU2013205085A1 (en) * 2013-02-22 2014-09-11 Leanne Therese Brown Shaver step
US9496050B2 (en) * 2013-05-22 2016-11-15 Micron Technology, Inc. Methods and apparatuses for stacked device testing
US9595349B2 (en) * 2015-06-25 2017-03-14 Intel Corporation Hardware apparatuses and methods to check data storage devices for transient faults
US9928924B2 (en) * 2015-12-15 2018-03-27 Qualcomm Incorporated Systems, methods, and computer programs for resolving dram defects
US10395748B2 (en) 2016-06-15 2019-08-27 Micron Technology, Inc. Shared error detection and correction memory

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6801471B2 (en) * 2002-02-19 2004-10-05 Infineon Technologies Ag Fuse concept and method of operation
US20080117696A1 (en) * 2006-11-20 2008-05-22 Chuan-Jen Chang Method for repairing defects in memory and related memory system
US20080141067A1 (en) * 2006-12-07 2008-06-12 Arm Limited Memory device and method of controlling access to such a memory device
US8924903B2 (en) * 2010-11-30 2014-12-30 Ps4 Luxco S.A.R.L. Semiconductor device having plural memory chip
US20140337573A1 (en) * 2013-05-08 2014-11-13 Qualcomm Incorporated Redirecting data from a defective data entry in memory to a redundant data entry prior to data access, and related systems and methods

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3472708A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10395748B2 (en) 2016-06-15 2019-08-27 Micron Technology, Inc. Shared error detection and correction memory
US11222708B2 (en) 2016-06-15 2022-01-11 Micron Technology, Inc. Shared error detection and correction memory

Also Published As

Publication number Publication date
EP3472708A4 (en) 2020-05-13
KR102399014B1 (en) 2022-05-17
US11222708B2 (en) 2022-01-11
EP3472708A1 (en) 2019-04-24
KR20200123279A (en) 2020-10-28
CN109313596B (en) 2022-06-17
KR20180138216A (en) 2018-12-28
KR102170322B1 (en) 2020-10-27
US20190362803A1 (en) 2019-11-28
US20170365356A1 (en) 2017-12-21
KR20210123403A (en) 2021-10-13
CN109313596A (en) 2019-02-05
CN114974393A (en) 2022-08-30
US10395748B2 (en) 2019-08-27

Similar Documents

Publication Publication Date Title
US11222708B2 (en) Shared error detection and correction memory
US10347355B2 (en) Device and method for repairing memory cell and memory system including the device
JP5875544B2 (en) Memory system
CN106548807B (en) Repair circuit, semiconductor device using the same, and semiconductor system
KR102117633B1 (en) Self repair device
US11113162B2 (en) Apparatuses and methods for repairing memory devices including a plurality of memory die and an interface
KR102076584B1 (en) Device and method of repairing memory cell and memory system including the same
US9401227B2 (en) Post package repair device
US20100271891A1 (en) Accessing Memory Cells in a Memory Circuit
KR20160148347A (en) Self repair device and method
CN115428086A (en) Apparatus and method for die replacement in stacked memory
US10790039B1 (en) Semiconductor device having a test circuit
US10460769B2 (en) Memory device including error detection circuit
US9418760B2 (en) Integrated circuit and method for testing semiconductor devices using the same
CN113362883A (en) Configurable Soft Post Package Repair (SPPR) scheme
US20210202477A1 (en) Three-dimensional device and manufacturing method thereof

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17813792

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20187036759

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017813792

Country of ref document: EP

Effective date: 20190115