Abstract: A stacked three-dimensional (3D) memory architecture is provided. An example stacked 3D memory architecture is included in a system and/or device, such as augmented reality glasses. Example augmented reality glasses include a camera, a 3D stacked memory, and a System-on-Chip (SoC). The 3D stacked memory is communicatively coupled with the camera and is configured to store image data captured by the camera. The 3D stacked memory includes a plurality of memory banks. The SoC is coupled with the 3D stacked memory. Additionally, the SoC is vertically stacked with the 3D stacked memory via a plurality of die-to-die interconnections between the SoC and the plurality of memory banks, includes a memory controller for accessing one or more memory banks of the plurality of memory banks, and is configured to process the image data stored in the 3D stacked memory.

Abstract: A headset includes a camera a 3D stacked memory configured to store image data captured by the camera, and a System-on-Chip (SoC) configured to process the image data stored in the 3D stacked memory. The 3D stacked memory includes a plurality of first drivers/receivers and a plurality of memory banks that are accessible in parallel. Each memory bank is accessible via a corresponding first driver/receiver. The SoC includes a memory controller with a plurality of second drivers/receivers. The plurality of the second drivers/receivers of the SoC are respectively connected to the plurality of the first drivers/receivers of the 3D stacked memory by a plurality of channels. The SoC and the 3D stacked memory are vertically stacked together. The plurality of the memory banks include at least eight memory banks.

Abstract: A method for pixel disaggregation can include forming a first layer of semiconductor material, wherein the first layer of semiconductor material includes memory and logic devices of a semiconductor device and is configured as a digital backplane that controls a plurality of analog devices of the semiconductor device. The method can additionally include forming a second layer of semiconductor material stacked atop the first layer of semiconductor material, wherein the second layer of semiconductor material includes the plurality of analog devices of the semiconductor device and is configured generate a per-pixel bias in response to control by the digital backplane. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A micro-light emitting diode (micro-LED) display backplane includes a plurality of macro-pixels. Each macro-pixel includes: a contiguous two-dimensional (2-D) array of bitcells storing display data bits for driving a set of micro-LEDs of a 2-D array of micro-LEDs; and drive circuits configured to generate, based on the display data bits stored in the contiguous 2-D array of bitcells, pulse-width modulated (PWM) drive signals for driving the set of micro-LEDs of the 2-D array of micro-LEDs. In one example, the plurality of macro-pixels is grouped into a plurality of sub-arrays, where each sub-array of the plurality of sub-arrays includes a set of macro-pixels and a local periphery circuit next to the set of macro-pixels. The local periphery circuit includes, for example, a buffer, a repeater, a clock gating circuit for gating an input clock signal to the sub-array, and/or a sub-array decoder for selecting the sub-array.

Abstract: A method for non-uniform memory access on three-dimensionally-stacked hybrid memory may include providing a logic die including a circuit and a memory. The method may additionally include providing a plurality of memory dies including an additional memory. The method may also include stacking the logic die and the plurality of memory dies three-dimensionally using face-to-face hybrid bonds that provide non-uniform access to the additional memory by the circuit. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: For small, high-resolution, light-emitting diode (LED) displays, such as for a near-eye display in an artificial-reality headset, LEDs are spaced closely together. A backplane can be used to drive an array of LEDs in an LED display. A plurality of interconnects electrically couple the backplane with the array of LEDs. The backplane can have a different coefficient of thermal expansion (CTE) than the array of LEDs. During bonding of the backplane to the array of LEDs, CTE mismatch can cause misalignment of bonding sites. The higher the bonding temperature, the greater the misalignment of bonding sites. Lower temperature bonding, using materials with lower melting or bonding temperatures, can be used to mitigate misalignment during bonding so that interconnects can be more closely spaced, which can allow LEDs to be more closely spaced, to enable a higher-resolution display.

Abstract: Described are techniques for generating a supply voltage for an SRAM array using power switching logic. The power switching logic can generate the supply voltage using a first supply rail (supplying a higher voltage) during an active state and using a second supply rail (supplying a lower voltage) during a deep retention state. In some examples, a sensing and recovery (SR) unit is provided to sense a decrease in the second voltage, for instance, during the deep retention state. The SR unit can generate an additional voltage that modifies the supply voltage to be higher than the decreased second voltage, thereby reducing droop and/or noise in the second supply rail. The power switching logic, SR unit, and SRAM array can be co-located or distributed across a computer system. For instance, the power switching logic, SR unit, and SRAM array can be embedded within a System on Chip integrated circuit.

Abstract: 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.

Abstract: Described are techniques for generating a supply voltage for an SRAM array using power switching logic. The power switching logic can generate the supply voltage using a first supply rail (supplying a higher voltage) during an active state and using a second supply rail (supplying a lower voltage) during a deep retention state. In some examples, a sensing and recovery (SR) unit is provided to sense a decrease in the second voltage, for instance, during the deep retention state. The SR unit can generate an additional voltage that modifies the supply voltage to be higher than the decreased second voltage, thereby reducing droop and/or noise in the second supply rail. The power switching logic, SR unit, and SRAM array can be co-located or distributed across a computer system. For instance, the power switching logic, SR unit, and SRAM array can be embedded within a System on Chip integrated circuit.

Abstract: In some embodiments, a system comprises a static random access memory (SRAM) device and a controller. The SRAM device comprises a bit cell array comprising a plurality of bit cells arranged in a plurality of rows and a plurality of columns, each column operatively coupled to a pair of bit lines, wherein the plurality of columns is arranged as a plurality of column groups each comprising a plurality of local columns. The SRAM device further comprises a plurality of column decoders, each associated with a column group of the plurality of column groups. In some embodiments, the controller may be configured to read the local columns included in the column group by, for a given local column, sensing a voltage difference on a corresponding pair of bit lines, in a rearranged sequential order that is different from a physical sequential order of the plurality of local columns.

Abstract: System on a Chip (SoC) integrated circuits are configured to reduce Static Random-Access Memory (SRAM) power leakage. For example, SoCs configured to reduce SRAM power leakage may form part of an artificial reality system including at least one head mounted display. Power switching logic on the SoC includes a first power gating transistor that supplies a first, higher voltage to an SRAM array when the SRAM array is in an active state, and a third power gating transistor that isolates a second power gating transistor from the first, higher voltage when the SRAM array is in the active state. The second power gating transistor further supplies a second, lower voltage to the SRAM array when the SRAM array is in a deep retention state, such that SRAM power leakage is reduced in the deep retention state.

Abstract: A micro-light emitting diode (micro-LED) display backplane includes a plurality of macro-pixels. Each macro-pixel includes: a contiguous two-dimensional (2-D) array of bitcells storing display data bits for driving a set of micro-LEDs of a 2-D array of micro-LEDs; and drive circuits configured to generate, based on the display data bits stored in the contiguous 2-D array of bitcells, pulse-width modulated (PWM) drive signals for driving the set of micro-LEDs of the 2-D array of micro-LEDs. In one example, the plurality of macro-pixels is grouped into a plurality of sub-arrays, where each sub-array of the plurality of sub-arrays includes a set of macro-pixels and a local periphery circuit next to the set of macro-pixels. The local periphery circuit includes, for example, a buffer, a repeater, a clock gating circuit for gating an input clock signal to the sub-array, and/or a sub-array decoder for selecting the sub-array.

Abstract: 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.

Abstract: In some embodiments, a system comprises a static random access memory (SRAM) device and a controller. The SRAM device comprises a bit cell array comprising a plurality of bit cells arranged in a plurality of rows and a plurality of columns, each column operatively coupled to a pair of bit lines, wherein the plurality of columns is arranged as a plurality of column groups each comprising a plurality of local columns. The SRAM device further comprises a plurality of column decoders, each associated with a column group of the plurality of column groups. In some embodiments, the controller may be configured to read the local columns included in the column group by, for a given local column, sensing a voltage difference on a corresponding pair of bit lines, in a rearranged sequential order that is different from a physical sequential order of the plurality of local columns.

Abstract: In some embodiments, an apparatus comprises: a static random access memory (SRAM) device. The SRAM device may have a bit cell array comprising a plurality of bit cells, the plurality of bit cells arranged in a plurality of rows and a plurality of columns, each column of the plurality of columns operatively coupled to a pair of bit lines. The apparatus may comprise a controller configured to: assert a word line associated with a row; perform a sequence of write operations while the word line remains asserted, each write operation corresponding to a bit cell associated with a different column of the plurality of columns and the row, wherein the word line has an elevated voltage relative to a non-elevated voltage during at least a portion of the sequence of write operations; and de-assert the word line after the sequence of write operations are performed.

Abstract: A display device includes a silicon wafer including digital circuits, a micro-light emitting diode (micro-LED) wafer including an array of micro-LEDs, and an indium-gallium-zinc-oxide (IGZO) layer between the silicon wafer and the micro-LED wafer and including analog circuits. The digital circuits are characterized by a first operating supply voltage and are configured to generate digital control signals based on digital display data of an image frame. The analog circuits are characterized by a second operating supply voltage higher than the first operating supply voltage. The analog circuits includes analog storage devices configured to storing analog signals, and transistors controlled by the digital control signals and the analog signals to generate drive currents for driving the array of micro-LEDs. The digital circuits on the silicon wafer or the analog circuits in the IGZO layer include level-shifting circuits at interfaces between the digital circuits and the analog circuits.

Abstract: A micro-light emitting diode (micro-LED) display backplane includes a plurality of macro-pixels. Each macro-pixel includes: a contiguous two-dimensional (2-D) array of bitcells storing display data bits for driving a set of micro-LEDs of a 2-D array of micro-LEDs; and drive circuits configured to generate, based on the display data bits stored in the contiguous 2-D array of bitcells, pulse-width modulated (PWM) drive signals for driving the set of micro-LEDs of the 2-D array of micro-LEDs. In one example, the plurality of macro-pixels is grouped into a plurality of sub-arrays, where each sub-array of the plurality of sub-arrays includes a set of macro-pixels and a local periphery circuit next to the set of macro-pixels. The local periphery circuit includes, for example, a buffer, a repeater, a clock gating circuit for gating an input clock signal to the sub-array, and/or a sub-array decoder for selecting the sub-array.

Abstract: System on a Chip (SoC) integrated circuits are configured to reduce Static Random-Access Memory (SRAM) power leakage. For example, SoCs configured to reduce SRAM power leakage may form part of an artificial reality system including at least one head mounted display. Power switching logic on the SoC includes a first power gating transistor that supplies a first, higher voltage to an SRAM array when the SRAM array is in an active state, and a third power gating transistor that isolates a second power gating transistor from the first, higher voltage when the SRAM array is in the active state. The second power gating transistor further supplies a second, lower voltage to the SRAM array when the SRAM array is in a deep retention state, such that SRAM power leakage is reduced in the deep retention state.

Abstract: For small, high-resolution, light-emitting diode (LED) displays, such as for a near-eye display in an artificial-reality headset, LEDs are spaced closely together. A backplane can be used to drive an array of LEDs in an LED display. A plurality of interconnects electrically couple the backplane with the array of LEDs. The backplane can have a different coefficient of thermal expansion (CTE) than the array of LEDs. During bonding of the backplane to the array of LEDs, CTE mismatch can cause misalignment of bonding sites. The higher the bonding temperature, the greater the misalignment of bonding sites. Lower temperature bonding, using materials with lower melting or bonding temperatures, can be used to mitigate misalignment during bonding so that interconnects can be more closely spaced, which can allow LEDs to be more closely spaced, to enable a higher-resolution display.

Abstract: For small, high-resolution, light-emitting diode (LED) displays, such as for a near-eye display in an artificial-reality headset, LEDs are spaced closely together. A backplane can be used to drive an array of LEDs in an LED display. A plurality of interconnects electrically couple the backplane with the array of LEDs. The backplane can have a different coefficient of thermal expansion (CTE) than the array of LEDs. During bonding of the backplane to the array of LEDs, CTE mismatch can cause misalignment of bonding sites. The higher the bonding temperature, the greater the misalignment of bonding sites. Lower temperature bonding, using materials with lower melting or bonding temperatures, can be used to mitigate misalignment during bonding so that interconnects can be more closely spaced, which can allow LEDs to be more closely spaced, to enable a higher-resolution display.