Embodiments of the disclosure are drawn to apparatuses and methods for scheduling targeted refreshes in a memory device. Memory cells in a memory device may be volatile and may need to be periodically refreshed as part of an auto-refresh operation. In addition, certain rows may experience faster degradation, and may need to undergo targeted refresh operations, where a specific targeted refresh address is provided and refreshed. The rate at which targeted refresh operations need to occur may be based on the rate at which memory cells are accessed. The memory device may monitor accesses to a bank of the memory, and may use a count of the accesses to determine if an auto-refresh address or a targeted refresh address will be refreshed.

Memory devices and systems with automatic background precondition upon powerup, and associated methods, are disclosed herein. In one embodiment, a memory device includes a memory array having a plurality of memory cells and a fuse array configured to store precondition data. The precondition data can identify a portion of the memory array, specify a predetermined precondition state, or a combination thereof. When the memory device powers on, the memory device can be configured to automatically retrieve the precondition data from the fuse array and/or to write memory cells in the portion of the memory array to the predetermined precondition state before executing an access command.

A semiconductor device includes a clock gating circuit and a control circuit. The clock gating circuit outputs a gated clock signal based on a clock signal. Transitions of the clock signal are output in the gated clock signal in response to a clock enable signal having an enable value and are disabled from being output in the gated clock signal in response to the clock enable signal having a disable value. The control circuit includes a first portion that operates based on the clock signal. The first portion sets the clock enable signal to the disable value in response to a disable control and sets the clock enable signal to the enable value in response to a wakeup control. The control circuit includes a second portion that operates based on the gated clock signal. The second portion provides the disable control to the first portion during an operation.

A semiconductor device includes a clock gating circuit and a control circuit. The clock gating circuit outputs a gated clock signal based on a clock signal. Transitions of the clock signal are output in the gated clock signal in response to a clock enable signal having an enable value and are disabled from being output in the gated clock signal in response to the clock enable signal having a disable value. The control circuit includes a first portion that operates based on the clock signal. The first portion sets the clock enable signal to the disable value in response to a disable control and sets the clock enable signal to the enable value in response to a wakeup control. The control circuit includes a second portion that operates based on the gated clock signal. The second portion provides the disable control to the first portion during an operation.

Embodiments herein may describe techniques for an integrated circuit including a FinFET transistor to be used as an antifuse element having a path through a fin area to couple a source electrode and a drain electrode after a programming operation is performed. A FinFET transistor may include a source electrode in contact with a source area, a drain electrode in contact with a drain area, a fin area including silicon and between the source area and the drain area, and a gate electrode above the fin area and above the substrate. After a programming operation is performed to apply a programming voltage between the source electrode and the drain electrode to generate a current between the source electrode, the fin area, and the drain electrode, a path may be formed through the fin area to couple the source electrode and the drain electrode. Other embodiments may be described and/or claimed.

Embodiments herein may describe techniques for an integrated circuit including a FinFET transistor to be used as an antifuse element having a path through a fin area to couple a source electrode and a drain electrode after a programming operation is performed. A FinFET transistor may include a source electrode in contact with a source area, a drain electrode in contact with a drain area, a fin area including silicon and between the source area and the drain area, and a gate electrode above the fin area and above the substrate. After a programming operation is performed to apply a programming voltage between the source electrode and the drain electrode to generate a current between the source electrode, the fin area, and the drain electrode, a path may be formed through the fin area to couple the source electrode and the drain electrode. Other embodiments may be described and/or claimed.

A multiply-accumulate (MAC) computation circuit includes: a source bit cell block configured to determine a MAC operation result of an input signal based on a plurality of source bit cells; a replica bit cell block comprising a plurality of replica bit cells corresponding to the plurality of source bit cells; and a readout circuit configured to read out a digital value of the MAC operation result using the replica bit cell block.

A multiply-accumulate (MAC) computation circuit includes: a source bit cell block configured to determine a MAC operation result of an input signal based on a plurality of source bit cells; a replica bit cell block comprising a plurality of replica bit cells corresponding to the plurality of source bit cells; and a readout circuit configured to read out a digital value of the MAC operation result using the replica bit cell block.

A neural network computation circuit that outputs output data according to a result of a multiply-accumulate operation between input data and connection weight coefficients, the neural network computation circuit includes computation units in each of which a memory element and a transistor are connected in series between data lines, a memory element and a transistor are connected in series between data lines, and gates of the transistors are connected to word lines. The connection weight coefficients are stored into the memory elements. A word line selection circuit places the word lines in a selection state or a non-selection state according to the input data. A determination circuit determines current values flowing in data lines to output output data. A current application circuit has a function of adjusting current values flowing in data lines, and adjusts connection weight coefficients without rewriting the memory elements.

A neural network computation circuit that outputs output data according to a result of a multiply-accumulate operation between input data and connection weight coefficients, the neural network computation circuit includes computation units in each of which a memory element and a transistor are connected in series between data lines, a memory element and a transistor are connected in series between data lines, and gates of the transistors are connected to word lines. The connection weight coefficients are stored into the memory elements. A word line selection circuit places the word lines in a selection state or a non-selection state according to the input data. A determination circuit determines current values flowing in data lines to output output data. A current application circuit has a function of adjusting current values flowing in data lines, and adjusts connection weight coefficients without rewriting the memory elements.