A memory architecture and a processing unit that incorporates the memory architecture and a systolic array. The memory architecture includes: memory array(s) with multi-port (MP) memory cells; first wordlines connected to the cells in each row; and, depending upon the embodiment, second wordlines connected to diagonals of cells or diagonals of sets of cells. Data from a data input matrix is written to the memory cells during first port write operations using the first wordlines and read out from the memory cells during second port read operations using the second wordlines. Due to the diagonal orientation of the second wordlines and due to additional features (e.g., additional rows of memory cells that store static zero data values or read data mask generators that generate read data masks), data read from the memory architecture and input directly into a systolic array is in the proper order, as specified by a data setup matrix.

To provide a semiconductor device with a novel structure. The semiconductor device includes an accelerator. The accelerator includes a first memory circuit, a second memory circuit, and an arithmetic circuit. The first memory circuit includes a first transistor. The second memory circuit includes a second transistor. Each of the first transistor and the second transistor includes a semiconductor layer including a metal oxide in a channel formation region. The arithmetic circuit includes a third transistor. The third transistor includes a semiconductor layer including silicon in a channel formation region. The first transistor and the second transistor are provided in different layers. The layer including the first transistor is provided over a layer including the third transistor. The layer including the second transistor is provided over the layer including the first transistor. The data retention characteristics of the first memory circuit are different from those of the second memory circuit.

The present disclosure provides a static random access memory (SRAM) cell. The SRAM cell includes first and second inverters cross-coupled for data storage, each inverter including at least one pull-up device and at least two pull-down devices; at least four pass gate devices configured with the two cross-coupled inverters; at least two ports coupled with the at least four pass-gate devices for reading and writing; a first contact feature contacting first two pull-down devices (PD-11 and PD-12) of the first inverter; and a second contact feature contacting second two pull-down devices (PD-21 and PD-22) of the second inerter.

The present disclosure provides a static random access memory (SRAM) cell. The SRAM cell includes first and second inverters cross-coupled for data storage, each inverter including at least one pull-up device and at least two pull-down devices; at least four pass gate devices configured with the two cross-coupled inverters; at least two ports coupled with the at least four pass-gate devices for reading and writing; a first contact feature contacting first two pull-down devices (PD-11 and PD-12) of the first inverter; and a second contact feature contacting second two pull-down devices (PD-21 and PD-22) of the second inerter.

A novel memory device is provided. The memory device includes a plurality of memory cells, and one memory cell includes a first transistor and a second transistor. One of a source and a drain of the first transistor is electrically connected to a gate of the second transistor through a node SN. Data written through the first transistor is retained at the node SN. When an OS transistor is used as the first transistor, formation of a storage capacitor is not needed. A region with a low dielectric constant is provided outside the memory cell, whereby noise from the outside is reduced and stable operation is achieved.

A 3D semiconductor device including: a first single crystal layer with first transistors; overlaid by a first metal layer; a second metal layer overlaying the first metal layer and being overlaid by a third metal layer; a logic gates including at least the first metal layer interconnecting the first transistors; second transistors disposed atop the third metal layer; third transistors disposed atop the second transistors; a top metal layer disposed atop the third transistors; and a memory array including word-lines, and at least four memory mini arrays, where each of the memory mini arrays includes at least four rows by four columns of memory cells, where each of the memory cells includes at least one of the second transistors or third transistors, sense amplifier circuit(s) for each of the memory mini arrays, the second metal layer provides a greater current carrying capacity than the third metal layer.

A 3D semiconductor device including: a first single crystal layer with first transistors; overlaid by a first metal layer; a second metal layer overlaying the first metal layer and being overlaid by a third metal layer; a logic gates including at least the first metal layer interconnecting the first transistors; second transistors disposed atop the third metal layer; third transistors disposed atop the second transistors; a top metal layer disposed atop the third transistors; and a memory array including word-lines, and at least four memory mini arrays, where each of the memory mini arrays includes at least four rows by four columns of memory cells, where each of the memory cells includes at least one of the second transistors or third transistors, sense amplifier circuit(s) for each of the memory mini arrays, the second metal layer provides a greater current carrying capacity than the third metal layer.

A dual-port memory cell includes first pull-up and pull-down transistors coupled at a first node between supply and reference voltage nodes, second pull-up and pull-down transistors coupled at a second node between the supply and reference voltage nodes, and first through fourth bit line landing pads in a metal layer. A first pass-gate transistor is coupled between the first bit line landing pad and the first node, a second pass-gate transistor is coupled between the second bit line landing pad and the second node, a third pass-gate transistor is coupled between the third bit line landing pad and the first node, and a fourth pass-gate transistor is coupled between the fourth bit line landing pad and the second node. The first node includes an interconnect between the first and second bit line landing pads, and the second node includes an interconnect between the third and fourth bit line landing pads.

A pseudo-triple-port memory is provided with read datapaths and write datapaths. The pseudo-triple-port memory includes a plurality of pseudo-triple-port bitcells, each pseudo-triple-port first bitcell having a first read port coupled to a first bit line, a second read port coupled to a second bit line, and a write port coupled to the first bit line and to the second bit line.

A floating body SRAM cell that is readily scalable for selection by a memory compiler for making memory arrays is provided. A method of selecting a floating body SRAM cell by a memory compiler for use in array design is provided.