Abstract: A memory unit with time domain edge delay accumulation for computing-in-memory applications is controlled by a first word line and a second word line. The memory unit includes at least one memory cell, at least one edge-delay cell multiplexor and at least one edge-delay cell. The at least one edge-delay cell includes a weight reader and a driver. The weight reader is configured to receive a weight and a multi-bit analog input voltage and generate a multi-bit voltage according to the weight and the multi-bit analog input voltage. The driver is connected to the weight reader and configured to receive an edge-input signal. The driver is configured to generate an edge-output signal having a delay time according to the edge-input signal and the multi-bit voltage. The delay time of the edge-output signal is positively correlated with the multi-bit analog input voltage multiplied by the weight.

Abstract: Disclosed herein is an apparatus that includes: a semiconductor substrate including first and second source regions coupled to a first power supply line and first and second drain regions coupled to a second power supply line, the first drain region being arranged between the first and second source regions, the second source region being arranged between the first and second drain regions; and gate electrodes including a first gate electrode arranged between the first source region and the first drain region, a second gate electrode arranged between the first drain region and the second source region, and a third gate electrode arranged between the second source region and the second drain region. The first and third gate electrodes are supplied with a first control signal. The second gate electrode is supplied with a second control signal.

Abstract: A programmable logic circuit includes multiple logic blocks that are connected communicatively, wherein multiple modules are reconfigured in any of the logic blocks, and wherein the modules include a first module that is being executed and a second module that is not being executed, and start of execution of the second module is delayed from a start time point of execution of the first module so as to obtain a state in which a first time at which the first module accesses a memory does not overlap a second time at which the second module accesses the memory.

Abstract: A circuit includes a memory array with memory cells arranged in a matrix of rows and columns, where each row includes a word line connected to the memory cells of the row, and each column includes a bit line connected to the memory cells of the column. Computational weights for an in-memory compute operation (IMCO) are stored in the memory cells. A word line control circuit simultaneously actuates word lines in response to input signals providing coefficient data for the IMCO by applying word line signal pulses. A column processing circuit connected to the bit lines processes analog signals developed on the bit lines in response to the simultaneous actuation of the word lines to generate multiply and accumulate output signals for the IMCO. Pulse widths of the signal pulses are modulated to compensate for cell drift. The IMCO further handles positive/negative calculation for the coefficient data and computational weights.

Abstract: A fast burst program sequence that reduces overall NAND flash programming time is disclosed. The burst program sequence includes maintaining a charge pump in an ON state and not fully discharging the WL/BLs at the conclusion of the programming phase of each program operation. As a result, the fast burst program sequence provides total program time savings over an existing cache program sequence by eliminating the full WL/BL discharge and charge pump reset that conventionally occurs after each program operation, which in turn, allows for the transfer of next page data from the page buffer to the data latches to be hidden within the program time of a prior/current program operation.

Abstract: Various implementations described herein relate to systems and methods for programming data, including determining a target row corresponding to a program command and setting row-based programming parameters for the target row using target physical device parameters of the target row and optimized programming parameters corresponding to the physical device parameters.

Abstract: First data is read out of a core storage array of a memory component over a first time interval constrained by data output bandwidth of the core storage array. After read out from the core storage array, the first data is output from the memory component over a second time interval that is shorter than the first time interval and that corresponds to a data transfer bandwidth greater than the data output bandwidth of the core storage array.

Abstract: A semiconductor device includes a memory core circuit configured to generate core data from bank data outputted by a bank or generate the core data from a dummy column address based on a read operation for the bank. The semiconductor device also includes a data control circuit configured to generate a switching signal from a bank active signal or a dummy bank address based on the read operation for the bank and and configured to control the output of the core data based on the switching signal.

Abstract: To provide a semiconductor device further reduced in size. A semiconductor device including: a multilayer wiring board one surface of which is provided with an external connection terminal; and a plurality of active components that are provided to be stacked inside the multilayer wiring board and are connected to the external connection terminal via a connection via. The plurality of active components include a first active component provided on another surface side that is opposite to the one surface, and a second active component that is provided closer to the one surface than the first active component is and has a smaller planar area than the first active component.

Abstract: A controller includes a central processing unit (CPU) configured to insert a latest received logical address, received together with a write command and data from a host, into a logical address list; a hotness determining circuit configured to assign a maximum weight to the latest received logical address, decrease weights of received logical addresses included in the logical address list by a decay factor, and sum weights of the received logical addresses having values, equal to a value of the latest received logical address, to determine hotness of the latest received logical address; and a parameter adjustment circuit decreasing a magnitude of the decay factor based on the repeatability index of the received logical addresses included in the logical address list, wherein the CPU is configured to control the memory device to store the data in one of the memory regions based on the hotness.

Abstract: An apparatus, a memory device, and a method for storing parameter codes with respect to asymmetric on-die-termination (ODT) are provided. The apparatus is connected to an external device via a signal line, and includes: an on-die termination (ODT) circuit set in a first ODT state; a plurality of signal pins, each of which is connected to the signal line; and an ODT control circuit configured to: identify whether a second ODT state of the external device corresponds to the first ODT state, and based on the apparatus being an asymmetric ODT in which the first ODT state and the second ODT state are different, provide an asymmetric ODT parameter code to the external device, and disable the ODT circuit when a signal is not transmitted through the signal line.

Abstract: A semiconductor memory device having a flexible refresh skip area includes a memory cell array including a plurality of rows to store data, a row decoder connected to the memory cell array, a refresh area storage unit to store a beginning address and an end address of a memory area that is to be refreshed in which the memory area that is to be refreshed does not include a refresh skip area having a size is selectively and/or adaptively changed, and a refresh control circuit connected to the row decoder and the refresh area storage unit. The refresh control circuit controls a refresh operation for the area that is to be refreshed and not for the refresh skip area.

Abstract: A processor-implemented data processing method includes: generating compressed data of first matrix data based on information of a distance between valid elements included in the first matrix data; fetching second matrix data based on the compressed data; and generating output matrix data based on the compressed data and the second matrix data.

Abstract: Methods and systems are disclosed for calibrating, by a memory interface system, an interface with dynamic random-access memory (DRAM) using a dynamically changing training clock. Techniques disclosed comprise receiving a system clock having a clock signal at a first pulse rate. Then, during the training of the interface, techniques disclosed comprise generating a training clock from the clock signal at the first pulse rate, the training clock having a clock signal at a second pulse rate, and sending, based on the generated training clock, command signals, including address data, to the DRAM.

Abstract: A memory system includes a first memory and a second memory that share common addresses received from a memory controller, wherein the first memory includes a first scrambling circuit suitable for scrambling a common address to generate a first scrambled address designating a word line to be activated in the first memory, and the second memory includes a second scrambling circuit suitable for scrambling the common address to generate a second scrambled address designating a word line to be activated in the second memory, and the first scrambling circuit and the second scrambling circuit perform a scrambling operation in such a manner that neighboring word lines, adjacent to a word line selected by a first common address, are selected a most in one memory among the first memory and the second memory by a second common address other than the first common address.

Abstract: In some embodiments, an integrated circuit may include a substrate and a memory array disposed on the substrate, where the memory array includes a plurality of discrete memory banks. The integrated circuit may also include a processing array disposed on the substrate, where the processing array includes a plurality of processor subunits, each one of the plurality of processor subunits being associated with one or more discrete memory banks among the plurality of discrete memory banks. The integrated circuit may also include a controller configured to implement at least one security measure with respect to an operation of the integrated circuit and take one or more remedial actions if the at least one security measure is triggered.

Abstract: A memory is provided that includes multiple memory banks, each one of the memory banks being associated with a read multiplexer. A first read multiplexer couples a first plurality of bit lines to a first sense node pair, and a second read multiplexer couples a second plurality of bit lines to a second sense node pair. A first sense amplifier is coupled to the first sense node pair. The second sense node pair may be coupled to the same sense amplifier or a different sense amplifier.

Abstract: Methods, systems, and devices for edgeless memory clusters are described. Systems, devices, and techniques are described for eliminating gaps between clusters by creating groups (e.g., domains) of clusters that are active at a given time, and using drivers within inactive clusters to perform array termination functions for abutting active clusters. Tiles on the edges of a cluster may have drivers that operate both for the cluster, and for a neighboring cluster, with circuits (e.g., a multiplexers) on the drivers to enable operations for both clusters.

Abstract: An interface circuit, a data transmission circuit and a memory are provided. The interface circuit includes a clock pad, data pads and input buffer circuits, where the clock pad and the data pads are arranged in the first row, and the M data pads are arranged on two sides of the clock pad, half of the M data pads being arranged on each side, where the M input buffer circuits are arranged in the second row and form an axis perpendicular to the first row with the data pads as reference, and the M input buffer circuits are arranged on two sides of the axis, half of the M input buffer circuits being arranged on each side, and where the distance between each input buffer circuit and the axis is smaller than the distance between the data pad corresponding to the input buffer circuit and the axis.

Abstract: A memory chip may include: a plurality of memory banks; a data storage configured to store access information indicative of access operations for one or more segments of the plurality of memory banks; and a refresh controller configured to perform a refresh operation of the one or more segments based, at least in part, on the stored access information.

Abstract: A pipe latch circuit may include first and second latching circuit groups. The first latching circuit group may control a latching operation and an output operation based on a plurality of pipe input control signals. The second latching circuit group may control a latching operation and an output operation based on the plurality of pipe input control signals and a plurality of pipe output control signals.

Abstract: A processor circuit sends a read request. A clock signal of the processor circuit corresponds to a first counting value. A memory circuit stores data and sends a data strobe signal in response to the read request. The data strobe signal corresponds to a second counting value. The processor circuit includes a selector circuit and a feedback circuit. The selector circuit selects and outputs a flag signal from a plurality of flag control signals according to the second counting value. The feedback circuit generates an enable signal according to a set signal associated with the first counting value, the flag signal associated with the second counting value, and a data strobe gate signal, and generates the data strobe gate signal according to the enable signal and the data strobe signal. The processor circuit reads the data according to the data strobe gate signal.

Abstract: A semiconductor memory device comprises: a substrate; a first conductive layer separated from the substrate in a first direction and extending in a second direction; a second and a third conductive layers separated from the substrate and the first conductive layer in the first direction and aligned in the second direction; a first semiconductor layer facing the first and the second conductive layers; a second semiconductor layer facing the first and the third conductive layers; a first and a second bit lines electrically connected to the first and the second semiconductor layers. At least some of operation parameters in the case of a certain operation being executed on a memory cell corresponding to the first conductive layer differ from at least some of operation parameters in the case of the certain operation being executed on a memory cell corresponding to the second conductive layer or the third conductive layer.

Abstract: A memory module including a memory array of storage transistors and a control circuit where the control circuit includes a memory interface for providing high bandwidth access to the memory array on serial data lanes. In some embodiments, the control circuit of a memory module includes multiple transceivers for connecting to serial data lanes. In one embodiment, the memory interface of a memory module configures some transceivers for host connection or for upstream connection to an upstream memory module and configures other transceivers for downstream connection to a downstream memory module. In other embodiments, a multi-module memory device is formed using multiple memory modules connected in a cascade configuration or in a star configuration to provide high bandwidth memory access to all memory locations of the multiple memory modules using the given number of serial data lanes of the host connection.

Abstract: Memory devices and methods of operating memory devices in which refresh management operations can be scheduled on an as-needed basis for those memory portions where activity (e.g., activations in excess of a predetermined threshold) warrants a refresh management operation are disclosed. In one embodiment, an apparatus comprises a memory including a memory location, and circuitry configured to determine a count corresponding to a number of activations at the memory location, to schedule a refresh management operation for the memory location in response to the count exceeding a first predetermined threshold, and to decrease the count by an amount corresponding to the first predetermined threshold in response to executing the scheduled refresh management operation. The circuitry may be further configured to disallow, in response to determining that the count has reached a maximum permitted value, further activations at the memory location until after the count has been decreased.

Abstract: A processing device in a memory sub-system detects an occurrence of a triggering event, determines respective levels of charge loss associated with a first representative wordline of a block of a memory device and with a second representative wordline of the block of the memory device, and determines whether a difference between the respective levels of charge loss satisfies a threshold criterion. Responsive to determining that the difference between the respective levels of charge loss satisfies the threshold criterion, the processing device further determines that the block is in a uniform charge loss state.

Abstract: Ranks of one or more memory modules can have variable data widths. As part of initializing a memory architecture, a memory controller can be configured to detect the data width of the ranks in the memory architecture. Based on the detected data widths of the ranks, the memory controller can synchronize the clocks and chip select signals of multiple ranks to thereby cause one or more memory chips in each of the multiple ranks to be connected to the shared data lines at the same time.

Abstract: An edge memory array mat with access lines that are split in half, and a bank of sense amplifiers formed in a region that separates the access line segment halves extending perpendicular to the access line segments. The sense amplifiers of the bank of sense amplifiers are coupled to opposing ends of a first subset of the half access lines pairs. The edge memory array mat further includes digit line (DL) jumpers or another structure configured to connect a second subset of the half access line pairs across the region occupied by the bank of sense amplifiers to form combined or extended access lines that extend to a bank of sense amplifiers coupled between the edge memory array mat and an inner memory array mat.

Abstract: Apparatuses for supplying power to a plurality of memory core chips are described. An example apparatus includes: a substrate, an interface chip on the substrate, and a plurality of memory core chips on the interface chip coupled to the interface chip via a plurality of electrodes. The plurality of memory core chips includes a first memory core chip, a second memory core chip, and a third memory core chip disposed between the second memory core chip and the interface chip. The first memory core chip and the third memory core chip are activated for data access while the second memory core chip disposed between the first memory core chip and the third memory core chip is deactivated for data access.

Abstract: Various embodiments of the present disclosure are directed towards a method for memory repair using a lookup table (LUT)-free dynamic memory allocation process. An array of memory cells having a plurality of rows and a plurality of columns is provided. Further, each memory cell of the array has multiple data states and a permanent state. One or more abnormal memory cells is/are identified in a row of the array and, in response to identifying an abnormal memory cell, the abnormal memory cell is set to the permanent state. The abnormal memory cells include failed memory cells and, in some embodiments, tail memory cells having marginal performance. During a read or write operation on the row, the one or more abnormal memory cells is/are identified by the permanent state and data is read from or written to a remainder of the memory cells while excluding the abnormal memory cell(s).

Abstract: Embodiments of a system and method for providing a flexible memory system are generally described herein. In some embodiments, a substrate is provided, wherein a stack of memory is coupled to the substrate. The stack of memory includes a number of vaults. A controller is also coupled to the substrate and includes a number of vault interface blocks coupled to the number of vaults of the stack of memory, wherein the number of vault interface blocks is less than the number of vaults.

Abstract: A memory is provided with a clock circuit configured to simultaneously assert a write multiplexer clock signal and a read multiplexer clock signal during a scan mode of operation. In the scan mode of operation, a scan in signal routes through a write multiplexer to a first bit line while the write multiplexer clock signal is asserted. Similarly, the scan in signal routes from the first bit line through a read multiplexer while the read multiplexer clock signal is asserted.

Abstract: Page data can be propagated sequentially from a section to the neighboring section, and from this section to subsequent section adjacent to it until a page register set is reached. In a described apparatus based on this page-data-copy scheme, access data from a page register (which is also used for storing the data accessed using the page-data-copy scheme) with a conditional read-access method in conjunction with an arithmetic unit can execute the arithmetic process of an AI system.

Abstract: The present disclosure provides a method for determining status of a fuse element of a memory device. The method includes providing the memory device including a first terminal and a second terminal and applying a first power signal on the first terminal of the semiconductor device. The memory device includes a configurable reference resistor unit electrically coupled to the fuse element. The method also includes obtaining an evaluation signal at the second terminal of the memory device and identifying the evaluation signal to determine whether the memory device is redundant. The configurable reference resistor unit includes a first resistor, a first transistor connected in parallel with the first resistor, and a first configurable unit connected to a gate of the first transistor. The first configurable unit is configured to generate a first configurable signal to turn on the first transistor.

Abstract: Disclosed herein is an apparatus that includes a first group including a plurality of first latch circuits coupled in series and a second group including a plurality of second latch circuits coupled in series. Each of the first latch circuits performs a latch operation in synchronization with a rise trigger signal. Each of the second latch circuits performs a latch operation in synchronization with a fall trigger signal. The rise and fall trigger signals are alternately activated every even clock cycles or every odd clock cycles. In response to a division ratio, first one or more of the first and second latch circuits are bypassed and second one or more of the first and second latch circuits are cyclically coupled.

Abstract: Methods and apparatuses for erasing data on a plurality of ferroelectric memory cells in a memory cell array in a memory apparatus are disclosed. An example apparatus includes: a memory cell array including a first plurality of word lines; a digit line; and a plurality of ferroelectric memory cells; a control circuit that provides a section select signal and a word line select signal to select a second plurality of word lines among the first plurality of word lines responsive to an address; and an address decoder that activates the second plurality of word lines. Each ferroelectric memory cell includes: a ferroelectric capacitor having a first terminal coupled to a cell plate node and a second terminal coupled to a selection circuit that couples the digit line to the second terminal responsive to a signal on a corresponding word line of the second plurality of word lines.

Abstract: Various implementations described herein relate to systems and methods for detecting address corruption when using a memory device to store and retrieve data, including but not limited to, reading combined information from a memory device, determining encoded data by de-combining address information from the combined information, and detecting address corruption by decoding the encoded data.

Abstract: The invention provides a multi-die stacked package memory and an output synchronization method thereof. The multi-die stacked package memory includes multiple dies (100), and the multiple dies (100) are stacked and packaged together to form a stacked package structure. The multiple dies (100) share a CS #pin, and the CS #pin is configured to turn on or turn off the stacked package structure. The multiple dies (100) also share an IO pin. Each die (100) is provided with a SYNC_PAD pin. The SYNC_PAD pins of the multiple dies (100) are electrically connected together, the SYNC_PAD pins are configured to judge whether the multiple dies (100) are all in an idle status or not. The multi-die stacked package memory and the output synchronization method thereof are simple in structure, easy to realize, stable and reliable.

Abstract: According to an embodiment, a semiconductor device includes a substrate, a connector, a volatile semiconductor memory element, multiple nonvolatile semiconductor memory elements, and a controller. A wiring pattern includes a signal line that is formed between the connector and the controller and that connects the connector to the controller. On the opposite side of the controller to the signal line, the multiple nonvolatile semiconductor memory elements are aligned along the longitudinal direction of the substrate.

Abstract: According to one embodiment, a semiconductor integrated circuit includes a memory region, a selection circuit, a setting register, and a setting circuit. The memory region stores plural kinds of setting conditions. The selection circuit selects a particular setting condition from the plural kinds of the setting conditions to read the particular setting condition out of the memory region. The setting register stores the particular setting condition read out of the memory region. The setting circuit refers to the particular setting condition stored in the setting register to set an operating condition.

Abstract: Systems, among other embodiments, include topologies (data and/or control/address information) between an integrated circuit buffer device (that may be coupled to a master, such as a memory controller) and a plurality of integrated circuit memory devices. For example, data may be provided between the plurality of integrated circuit memory devices and the integrated circuit buffer device using separate segmented (or point-to-point link) signal paths in response to control/address information provided from the integrated circuit buffer device to the plurality of integrated circuit buffer devices using a single fly-by (or bus) signal path. An integrated circuit buffer device enables configurable effective memory organization of the plurality of integrated circuit memory devices. The memory organization represented by the integrated circuit buffer device to a memory controller may be different than the actual memory organization behind or coupled to the integrated circuit buffer device.

Abstract: Voltages loaded onto the bit lines in a first CA section of a memory array can be latched by enabling the BLSA between the first section and a second section adjacent to the first section causing latched voltages to propagate to bit lines in the second section. Voltages propagated to the bit lines in the second section using the latches between the second section and a third section. Voltages can be propagated sequentially from section to subsequent adjacent section until a target location is reached. The scheme can be applied as a method of page-data write access in a memory chip, of which page data can be propagated sequentially from section to subsequent adjacent section until a target location is reached, and then, activating a word line in a section of the memory comprising the target location to write voltages to the memory cells at the target location.

Abstract: A layout for a 6T SRAM cell array is disclosed. The layout doubles the number of bits per bit cell in the array by implementing dual pairs of bitlines spanning bit cell columns in the array. Alternating connections (e.g., alternating vias) may be provided for wordline access to the bitlines in the layout. Alternating the connections may reduce RC delay in the layout.

Abstract: An anti-fuse memory circuit includes: a memory array including multiple anti-fuse memory cells; bit lines, each connected to the anti-fuse memory cells arranged in extension direction of the bit line, each anti-fuse memory cell being electrically connected to respective one of bit lines through first switch transistor; word lines, each connected to first switch transistors arranged in extension direction of word line; a second switch transistor connects one of the bit lines to transmission wire; a reading circuit, having first input terminal connected to the transmission wire, second input terminal for receiving reference voltage, and sampling input terminal for receiving sampling signal; and a signal generation circuit for generating sampling signal according to precharge voltage and precharge signal, where precharge signal is used for instructing to precharge transmission wire to precharge voltage, and delay duration between sampling signal and precharge signal is positively correlated with voltage amplitud

Abstract: A non-volatile memory device includes a plurality of memory cells arranged in a matrix, a plurality of word lines extended in a row direction, and a plurality of bit lines extended in a column direction. Each of the memory cells is coupled to one of the word lines and one of the bit lines. The memory device further includes a word-line control circuit coupled to and configured to control the word lines, a first bit-line control circuit configured to control the bit lines and sense the memory cells in a digital mode, and a second bit-line control circuit configured to bias the bit lines and sense the memory cells in an analog mode. The first bit-line control circuit is coupled to a first end of each of the bit lines. The second bit-line control circuit is coupled to a second end of each of the bit lines.

Abstract: A memory system includes: a memory device including at least one bank; and a memory controller suitable for: dividing the bank into a plurality of sub-regions according to an active address, generating an aging signal for the bank based on a plurality of counting values generated by counting a number of inputs of an active command for each of the sub-regions, and providing the active command, the active address, the aging signal, and a target refresh command, wherein the memory device is suitable for: generating a sampling address by sampling the active address according to the active command, and performing a target refresh operation on a word line corresponding to the sampling address according to the target refresh command while adjusting a refresh period of the bank according to the aging signal.

Abstract: Systems and method are provided for a memory circuit. In embodiments, the circuit includes a plurality of memory cells corresponding to a word of data and a global write word line. A plurality of local write lines are connected to a subset of the plurality of memory cells of the word of data. Selection logic is configured to activate a particular subset of memory cells for writing via a particular local write line based on a signal on the global write word line and a selection signal associated with the particular subset of memory cells.

Abstract: Implementations described and claimed herein provide a high-capacity, high-bandwidth scalable storage device. The scalable storage device includes a layer stack including at least one memory layer and at least one optical control layer positioned adjacent to the memory layer. The memory layer includes a plurality of memory cells and the optical control layer is adapted to receive optically-encoded read/write signals and to effect read and write operations to the plurality of memory cells through an electrical interface.

Abstract: Disclosed herein are related to a memory system including unit storage circuits. In one aspect, each of the unit storage circuits abuts an adjacent one of the unit storage circuits. In one aspect, each of the unit storage circuits includes a first group of memory cells, a second group of memory cells, a first sub-word line driver to apply a first control signal to the first group of memory cells through a first sub-word line extending along a direction, and a second sub-word line driver to apply a second control signal to the second group of memory cells through a second sub-word line extending along the direction. In one aspect, the memory system includes a common word line driver abutting one of the unit storage circuits and configured to apply a common control signal to the unit storage circuits through a word line extending along the direction.

Abstract: Memory operations using compound memory commands, including: receiving, by a memory module, a compound memory command indicating one or more operations to be applied to each portion of a plurality of portions of contiguous memory in the memory module; generating, based on the compound memory command, a plurality of memory commands to apply the one or more operations to each portion of the plurality of portions of contiguous memory; and executing the plurality of memory commands.