Abstract: A compute-in-memory device may include a Booth encoder configured to receive at least one input of first bits, a Booth decoder configured to receive at least one weight of second bits and to output a plurality of partial products of the at least one input and the at least one weight, an adder configured to add a first partial product of the plurality of the partial products and a second partial product of the plurality of partial products before the Booth decoder generates a third partial product of the plurality of the partial products and to generate a plurality of sums of partial products, and a carry-lookahead adder configured to add the plurality of sums of partial products and to generate a final sum.

Abstract: A device includes first and second gate electrodes, a word line and a first metal island. The first gate electrode corresponds to transistors of a memory cell. The second gate electrode is separated from the first gate electrode and corresponds to the transistors. The word line is coupled to the memory cell and located between the first and the second gate electrodes. The first metal island is configured to couple a first power supply to the memory cell. A first boundary of the first metal island is located between first and second boundaries of the first gate electrode and is located between first and second boundaries of the word line, and each of the first boundary of the first gate electrode and the first boundary of the word line is located between first and second boundaries of the first metal island.

Abstract: A memory cell includes a write port and a read port. The write port includes two cross-coupled inverters that form a storage unit. The cross-coupled inverters are connected between a first power source signal line and a second power source signal line. The write port also includes a first local interconnect line in an interconnect layer that is connected to the second power source signal line. The read port includes a transistor that is connected to the storage unit in the write port and to the second power source signal line, and a second local interconnect line in the interconnect layer that is connected to the second power source signal line. The second local interconnect line in the read port is separate from the first local interconnect line in the write port.

Abstract: A memory device includes a first bit cell, a second bit cell, a first word line and a second word line. A first boundary of the second bit cell is adjacent with a first boundary of the first bit cell. The first word line is coupled to the first bit cell. The second word line is coupled to the second bit cell. A first segment of the first word line is overlapped with the first boundary of the second bit cell in a plan view, and a first segment of the second word line is overlapped with a second boundary of the second bit cell in the plan view.

Abstract: An in-memory computing circuit is provided. The in-memory computing circuit includes a core die, a plurality of conductive pillars, and a plurality of memory dies. The plurality of memory dies are coupled to the core die through the plurality of conductive pillars and are configured to implement computing operation. The plurality of memory dies includes at least one of the memory dies disposed on a bottommost memory die of the plurality of memory dies. The plurality of memory dies receives an input data from the core die through a common input terminal of the core die.

Abstract: A memory device is disclosed. The memory device includes a first program line and a second program line. A first portion of the first program line is formed in a first conductive layer, and a second portion of the first program line is formed in a second conductive layer above the first conductive layer. A first portion of the second program line is formed in the first conductive layer, and a second portion of the second program line is formed in a third conductive layer above the second conductive layer. A width of at least one of the second portion of the first program line or the second portion of the second program line is different from a width of at least one of the first portion of the first program line or the first portion of the second program line. A method is also disclosed herein.

Abstract: A semiconductor memory device includes: a local write bit (LWB) line; a local write bit_bar (LWB_bar) line; a global write bit (GWB) line; a global write bit_bar (GWBL_bar) line; a column of segments, each segment including bit cells that are connected correspondingly between the LWB and LWB_bar lines; and a distributed write driving arrangement including a global write driver and local write drivers included correspondingly in the segments; and the global write driver including a first equalizer circuit, arranged in a switched-coupling between the LWB line and the LWB_bar line, and arranged in a control-coupling with respect to signals correspondingly on the GWB line and the GWB_bar line, and the global write driver and the local write drivers each including first inversion couplings (coupled in parallel between the GWB line and the LWB line) and second inversion couplings (coupled in parallel between the GWB_bar line and the LWB_bar line).

Abstract: A device includes a first memory cell, a second memory cell, a first logic element, a second logic element, and a third logic element. The first memory cell is configured to store a first bit at a first node, and the second memory cell is configured to store a second bit at a second node. The first logic element includes a first node input terminal coupled to the first node, the second logic element includes a second node input terminal coupled to the second node, and the third logic element includes a first input terminal coupled to a first output terminal of the first logic element and a second input terminal coupled to a second output terminal of the second logic element.

Abstract: A device is disclosed that includes a fin structure disposed below a first metal layer, extending along a column direction, and corresponding to at least one transistor of a memory bit cell, a word line disposed in the first metal layer and extending along a row direction, a first metal island disposed in the first metal layer separated from the word line, and a first connection metal line disposed in a second metal layer above the first metal layer, extending along the column direction, and configured to couple a power supply through the first metal island to the fin structure. In a layout view, the first connection metal line is separated from the fin structure, and the fin structure crosses over the word line and the first metal island. A method is also disclosed herein.

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: A semiconductor chip is provided. The semiconductor chip includes a SRAM cell, a logic cell, a signal line and a ground line. The SRAM cell includes a storage transmission gate, a read transmission gate and a latch circuit. The latch circuit is serially connected between the storage and read transmission gates, and includes a first inverter, a second inverter and a transmission gate connected to an output of the first inverter, an input of the second inverter and an output of the storage transmission gate. The logic cell disposed aside the SRAM cell is connected with the SRAM cell by first and second active structures. The signal and ground lines extend at opposite sides of the SRAM and logic cells, and are substantially parallel with the first and second active structures. The SRAM and logic cells are disposed between and electrically connected to the signal and ground lines.

Abstract: A method of forming a memory circuit includes generating a layout design of the memory circuit, and manufacturing the memory circuit based on the layout design. The generating of the layout design includes generating a first active region layout pattern corresponding to fabricating a first active region of a first pull down transistor, generating a second active region layout pattern corresponding to fabricating a second active region of a first pass gate transistor, and generating a first metal contact layout pattern corresponding to fabricating a first metal contact. The first metal contact layout pattern overlaps the cell boundary of the memory circuit and the first active region layout pattern. The first metal contact electrically coupled to a source of the first pull down transistor. The memory circuit being a four transistor (4T) memory cell including a first and second pass gate transistor, and a first and second pull down transistor.

Abstract: A memory device and an operating method for computing-in-memory (CIM) are provided. The memory device for CIM comprises a plurality of memory banks and a global multiply accumulate (MAC) circuit. Each of the memory banks comprises a first memory array, a first latch circuit, a second latch circuit and a local MAC circuit. The first latch circuit latches a first data from the first memory array in a first read cycle. The second latch circuit latches a second data from the first memory array in a second read cycle. The local MAC circuit performs a first stage CIM operation on a first latched data latched in the first latch circuit and the second latched data latched in the second latch circuit to provide a first stage CIM result. The global MAC circuit performs a second stage CIM operation on a plurality of first stage CIM results from the memory banks.

Abstract: A device including: a first adder having first adder inputs and first adder outputs; a first register having first register inputs and first register outputs, the first register inputs coupled to the first adder outputs; a second register having second register inputs and second register outputs, the second register inputs coupled to the first adder outputs; and a second adder having second adder inputs and second adder outputs and configured to receive register output signals from the first register outputs and the second register outputs. Wherein, the first adder is configured to calculate a first sum of a first input value and a second input value, and the first register is configured to store the first sum, and the first adder is configured to calculate a second sum of a third input value and a fourth input value, and the second register is configured to store the second sum.

Abstract: A memory device has a memory array of a plurality of memory cells arranged in a plurality of columns and a plurality of rows. The memory cells in each of the plurality of columns include first memory cells and second memory cells alternately arranged along a column direction of the plurality of columns. A first computation circuit is coupled to the first memory cells in each of the plurality of columns, and is configured to generate first output data corresponding to a first computation performed on first weight data stored in the first memory cells. A second computation circuit is coupled to the second memory cells in each of the plurality of columns, and is configured to generate second output data corresponding to a second computation performed on second weight data stored in the second memory cells.

Abstract: An SRAM includes multiple memory cells, each memory cell includes a data storage unit; a data I/O control adapted to input data to, and output data from, a data line; and multiple access controls respectively connected to at least two access control lines (WL's) and adapted to enable and disable the data input and output from the at least two WL's. The access controls are configured to permit data input only when both WL's are in their respective states that permit data input. A method of writing to a group of SRAM cells include sending a first write-enable signal to the cells via a first WL, sending a group of respective second write-enable signals to the respective cells, and, for each of the cells, preventing writing data to the cell if either of the first write-enable signal and respective second write enable signal is in a disable-state.

Abstract: A charge sharing scheme is used to mitigate the variations in cell currents in order to achieve higher accuracy for CIM computing. In some embodiments, a capacitor is associated with each SRAM cell, and the capacitors associated with all SRAM cells in a column are included in averaging the RBL current. In some embodiments, a memory unit associated to an RBL in a CIM device includes a storage element adapted to store a weight, a first switch device connected to the storage element and adapted to be controlled by an input signal and generate a product signal having a magnitude indicative of the product of the input signal and the stored weight. The memory unit further includes a capacitor adapted to receive the product signal and store an amount of charge corresponding to the magnitude of the product signal. The memory unit further include a second switch device adapted to transfer the charge on the capacitor to the RBL.

Abstract: Header circuitry for a memory device includes multiple backside power rails that form distinct voltage sources for a plurality of switching devices in the header circuitry. The header circuitry includes at least one region of a first conductivity type. A first section in the first region includes one backside power rail (BPR) that forms a first voltage source that provides a first voltage. A second section in the same first region includes another BPR that forms a second voltage source that provides a second voltage that is different from the first voltage.

Abstract: A memory device includes a conductive segment, first and second rows of memory cells. The conductive segment receives a first reference voltage signal. The first row of memory cells is coupled to a first word line. The second row of memory cells is coupled to a second word line. The first row of memory cells includes first and second memory cells. The first memory cell is coupled to the conductive segment to receive the first reference voltage signal. The second row of memory cells includes third and fourth memory cells. The third memory cell is coupled to the conductive segment to receive the first reference voltage signal. The first and third memory cells share the conductive segment, and the third memory cell is arranged between the first and second memory cells. The second memory cell is arranged between the third and fourth memory cells.

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.