A monotonic counter stores N binary words representing a value in N memory cells. When i memory cells of consecutive ranks between k modulo N and k+i modulo N each represent a value complementary to a null value, the counter is incremented by erasing a value of a memory cell of rank k+i+1 modulo N. When i+1 memory cells of consecutive ranks between k+1 modulo N and k+i+1 modulo N each represent the value complementary to the null value, the counter is incremented by incrementing a value of a memory cell of rank k modulo N by two step sizes and storing a result in a memory cell of rank k+1 modulo N, wherein, N is an integer greater than or equal to five, k is an integer, and i is an integer between 2 and N−3.

A monotonic counter stores N binary words representing a value in N memory cells. When i memory cells of consecutive ranks between k modulo N and k+i modulo N each represent a value complementary to a null value, the counter is incremented by erasing a value of a memory cell of rank k+i+1 modulo N. When i+1 memory cells of consecutive ranks between k+1 modulo N and k+i+1 modulo N each represent the value complementary to the null value, the counter is incremented by incrementing a value of a memory cell of rank k modulo N by two step sizes and storing a result in a memory cell of rank k+1 modulo N, wherein, N is an integer greater than or equal to five, k is an integer, and i is an integer between 2 and N−3.

A memory device according to the present invention comprises: a memory cell array in which memory cells are connected to wordlines and bitlines in a matrix form; and a control circuit for programming the memory cells or controlling a read operation, according to a start address, a burst length, a latency length, and a program or read command which are transmitted from a host, wherein the control circuit may comprise: a pulse generation unit for generating register pulses and counter pulses in synchronization with an operation clock; and a counter that sets the start address in synchronization with the register pulses, counts the number of counter pulses corresponding to the sum of the latency length and the burst length, and increases an address from the start address to the sum of the start address and the burst length.

A memory device includes a memory cell array and a peripheral circuit. The memory cell array includes a plurality of memory regions each identified by a row address and a column address. The peripheral circuit accesses the memory cell array by performing, based on an address, a burst length and a burst address gap provided from a memory controller, a burst operation supporting a variable burst address gap. The burst address gap is a numerical difference between adjacent column addresses, on which the burst operation is to be performed.

A memory device includes a memory cell array and a peripheral circuit. The memory cell array includes a plurality of memory regions each identified by a row address and a column address. The peripheral circuit accesses the memory cell array by performing, based on an address, a burst length and a burst address gap provided from a memory controller, a burst operation supporting a variable burst address gap. The burst address gap is a numerical difference between adjacent column addresses, on which the burst operation is to be performed.

In some embodiments, an in-memory-computing SRAM macro based on capacitive-coupling computing (C3) (which is referred to herein as “C3SRAM”) is provided. In some embodiments, a C3SRAM macro can support array-level fully parallel computation, multi-bit outputs, and configurable multi-bit inputs. The macro can include circuits embedded in bitcells and peripherals to perform hardware acceleration for neural networks with binarized weights and activations in some embodiments. In some embodiments, the macro utilizes analog-mixed-signal capacitive-coupling computing to evaluate the main computations of binary neural networks, binary-multiply-and-accumulate operations. Without needing to access the stored weights by individual row, the macro can assert all of its rows simultaneously and form an analog voltage at the read bitline node through capacitive voltage division, in some embodiments. With one analog-to-digital converter (ADC) per column, the macro cab realize fully parallel vector-matrix multiplication in a single cycle in accordance with some embodiments.

A semiconductor device includes: a memory cell array including a plurality of memory cells; a data input/output circuit suitable for outputting data provided from the memory cell array in response to a couple of data output control signals; and a data output control circuit suitable for generating a couple of latch read enable signals and a couple of data output control timing signals based on a couple of complementary read enable signals, an internal enable signal and warming-up cycle information indicating different warming-up cycles, and outputting, according to the couple of data output control timing signals, the couple of data output control signals using the couple of latch read enable signals, one or more pulses of each of which are masked according to the warming-up cycle information.

A semiconductor device includes: a memory cell array including a plurality of memory cells; a data input/output circuit suitable for outputting data provided from the memory cell array in response to a couple of data output control signals; and a data output control circuit suitable for generating a couple of latch read enable signals and a couple of data output control timing signals based on a couple of complementary read enable signals, an internal enable signal and warming-up cycle information indicating different warming-up cycles, and outputting, according to the couple of data output control timing signals, the couple of data output control signals using the couple of latch read enable signals, one or more pulses of each of which are masked according to the warming-up cycle information.

An SRAM controller for performing sequential accesses using internal ports that operate concurrently on different rows. Each internal port includes a row address strobe (RAS) timer that generates clock signals controlling the timing of operations during a RAS phase in which word line decoding is performed once for a group of bit cells being accessed. The RAS phase can involve additional conditioning operations, such as precharging of local bits lines associated with the group of bit cells. The RAS phase is followed by an input/output (IO) phase in which individual bit cells are accessed in sequential address order using a column select signal generated by an IO timer. The RAS phase of a first internal port can be at least partially overlapped by the IO phase of a second internal port to hide the RAS latency of the first internal port. The IO timer can be shared among internal ports.

A storage device includes a memory including a plurality of regions arranged along a first axis and a second axis orthogonal to each other, each of the plurality of regions belonging to one of first groups and one of second groups; and a controller configured to, when a programmed and weak region exists, put into a scan list on the basis of a weak list, a programmed and weak sub-region included in the programmed and weak region among the plurality of regions, put into the scan list, a first programmed and adjacent sub-region in a first programmed and adjacent region selected according to a second axis expansion order among the plurality of regions, and put into the scan list, a second programmed and adjacent sub-region in a second programmed and adjacent region selected according to a first axis expansion order among the plurality of regions.