Abstract: A multiport SRAM has an array of cells, a first port, and a second port. During a period of different row addresses for the ports, the first port uses first word lines and first bit lines. The second port uses second word lines and second bit lines. In response to the second port switching to the same address as the first port to make a row match, the second port and the first port use the first plurality of word lines, but the first port uses the first plurality of bit lines and the second port uses the second plurality of bit lines. If the row match is removed by the first port changing row addresses, a correlation swap is performed so that the first port performs accesses using the second word lines and bit lines and the second port performs accesses using the first word lines and bit lines.

Abstract: A method and apparatus are provided for error correction of a memory by using a first memory (18) and second memory (14) to perform error correction code (ECC) processing on data retrieved from the first memory and to use status control bits (35-37) in the second memory to detect and manage hard and soft errors identified by the ECC processing.

Abstract: A memory cell includes a first bi-directional resistive element having a cathode coupled to a first power rail and an anode coupled to an internal node, a second bi-directional resistive element having a cathode coupled to the internal node and an anode coupled to a second power rail, and a first transistor having a control electrode coupled to the internal node, a first current electrode coupled to a first bitline, and a second current electrode coupled to a third power rail.

Abstract: Structures for substituting single bits in an array may include a first array having a plurality of word lines, and for each of the plurality of word lines a memory operable to store a bit substitution column value, and a first data output line operable to communicate the bit substitution column value to a write data shifter. The bit substitution column value may be associated with a bit substitution column in a second array, and the write data shifter may be operable to substitute the bit by shifting data to a redundancy column in the first array.

Abstract: The dies of a stacked die IC are tested and, in response to detection of a defect at one of the dies, the type of defect is identified. If the defect is identified as a defective module repairable at the die itself, a redundant module of the die is used to replace the functionality of the defective module. If the defect is identified as one that is not repairable, a replacement die in the die stack is used to replace the functionality of the defective die.

Abstract: A high density, low power, high performance information system, method and apparatus are described in which perpendicularly oriented processor and memory die stacks (130, 140, 150, 160, 170) include integrated deflectable MEMS optical beam waveguides (e.g., 190) at each die edge (e.g., 151) to provide optical communications (184) in and between die stacks by using a beam control method and circuit to maintain and adjust alignment over time by calibrating and updating X and Y counter values stored in deflection registers (541-542) to control DAC circuitry (546, 548) which generates and supplies deflection voltages to charging capacitors (551, 552) connected to deflection electrodes (195-197) positioned on and around each MEMS optical beam waveguide (193-194) to provide two-dimensional alignment and controlled feedback to adjust beam alignment and establish optical communication links between die stacks.

Abstract: A method and apparatus are provided for error correction of a memory by using a first memory (18), second memory (14), and redundant memory (19) to perform error correction code (ECC) processing on data retrieved from the first memory (18) by using the redundant memory (19) to replace entries in the second memory (14) having repeat addresses, thereby freeing entries in the second memory (14) for use in detecting and managing errors identified by the ECC processing.

Abstract: A circuit for determining a threshold indication of temperature with respect to a threshold temperature. The circuit includes a timer circuit and a temperature sensor circuit having an counter whose output has a relationship to temperature. At the end of a period determined by the timer circuit, a comparator circuit compares the count of the counter with an indication of the threshold temperature to determine a state of the threshold indication. In response to a change in state of the threshold indication, the circuit changes one of the count time or the counter output's relationship to temperature to provide a hysteresis for the threshold indication.

Abstract: A charge pump system includes a comparator having a first input coupled to a first reference voltage, a second input coupled to a feedback signal and an output coupled to control operation of a voltage controlled oscillator. The feedback signal is coupled to an output of the charge pump system. An amplifier has a first input coupled to a second reference voltage, a second input coupled to the feedback signal, and an output coupled as input to the voltage controlled oscillator. A gain of the amplifier is lower than a gain of the comparator.

Abstract: A memory device includes a storage unit formed using a substrate, a true bit line BL0 for carrying a bit of data, and a complementary bit line for carrying the bit of data carried by the first true bit line in complementary form. The true bit line is coupled to the storage unit and runs laterally over the substrate. The true bit line and the complementary bit line are adjacent to each other and are vertically stacked above the substrate.

Abstract: An SRAM bit cell comprises a first inverter including a PMOS transistor and an NMOS transistor, and a second inverter including a PMOS transistor and an NMOS transistor. The first and second inverters are cross-coupled to each other. A plurality of pass transistors couple the inverters to bit lines. Approximately one-half of a supply voltage is provided to the bit lines during pre-charge operations.

Abstract: An electronic assembly includes a processor die assembly, a first die assembly, and a second die assembly. The first die assembly is positioned on a first side of the processor die assembly. The second die assembly is positioned on a second side of the processor die assembly opposite the first side of the processor die assembly. Through-die vias couple the first and second die assemblies to the processor die assembly.

Abstract: A method of manufacturing a microelectronic assembly (100) and a microelectronic device (4100) that include a stacked structure (101). The stacked structure includes a heat spreader (104), at least one die (106) thermally coupled to at least a portion of one side of the heat spreader, at least one other die (108) thermal coupled to at least a portion of an opposite side of the heat spreader, at least one opening (401) in the heat spreader located in a region of between the two die, an insulator (603) disposed in the at least one opening, and electrically conductive material (1308, 1406) in an insulated hole (705) in the insulator. The heat spreader allows electrical communication between the two die through the opening while the insulator isolates the electrically conductive material and the heat spreader from each other.

Abstract: A high density, low power, high performance information system, method and apparatus are described in which a laser source (213) on a first die (210) generates a source light beam of unmodulated monochromatic coherent light (281) for distribution via optical beam routing structures (e.g., 214/214a, 224/224a, 234/234a) to a plurality of receiving die (220, 230), each of which includes its own modulator (e.g., 223, 233) for optically receiving at least a portion of the source light beam (281a, 281b) from the first die and generating therefrom an output source light beam of modulated monochromatic coherent light (291, 292) which is encoded at said modulator in response to electrical signal information.

Abstract: A high density, low power, high performance information system, method and apparatus are described in which perpendicularly oriented processor and memory die stacks (130, 140, 150, 160, 170) include integrated deflectable MEMS optical beam waveguides (e.g., 190) at each die edge to provide optical communications (182-185) in and between die stacks by supplying deflection voltages to a plurality of deflection electrodes (195-197) positioned on and around each MEMS optical beam waveguide (193-194) to provide two-dimensional alignment and controlled feedback to adjust beam alignment and establish optical communication links between die stacks.

Abstract: A semiconductor device comprising a substrate, a power bus, a heat source circuit, a heat sensitive circuit, and a plurality of electrically and thermally conductive through-silicon-vias (TSVs) in the substrate. The TSVs are electrically coupled to the power bus and positioned between the heat source circuit and the heat sensitive circuit to absorb heat from the heat source circuit.

Abstract: A low profile strip dual in-line memory module (200) includes a passive interposer support structure (90) with patterned openings (91-97) formed between opposing top and bottom surfaces, a plurality of memory chips (D1-D8) attached to the top and bottom surfaces, and vertical solder ball conductors (98) extending through the patterned openings to electrically connect the plurality of memory chips, where each memory chip has an attachment surface facing the passive interposer structure and a patterned array of horizontal conductors (e.g., 82-86) formed on the attachment surface with contact pads electrically connected to the plurality of vertical conductors to define at least one bus conductor that is electrically connected to each memory die in the first and second plurality of memory die.

Abstract: Edge coupling of semiconductor dies. In some embodiments, a semiconductor device may include a first semiconductor die, a second semiconductor die disposed in a face-to-face configuration with respect to the first semiconductor die, and an interposer arranged between the first semiconductor and second semiconductor dies, the interposer having an edge detent configured to allow an electrical coupling between the first and second semiconductor dies. In other embodiments, a method may include coupling a first semiconductor die to a surface of an interposer where an edge of the interposer includes detents and the first semiconductor die includes a first pad aligned with a first detent, coupling a second semiconductor die to an opposite surface of the interposer where the first and second semiconductor dies are in a face-to-face configuration and the second semiconductor die includes a second pad aligned with a second detent, and coupling the first and second pads together.

Abstract: A buffer circuit comprising a capacitor, the capacitor comprising a first terminal and a second terminal, an input signal being coupled to the first terminal; a first buffer stage coupled to the second terminal of the capacitor such that the input signal is capacitively coupled to the first buffer stage; and, a pulse generator coupled to control the first buffer stage, the pulse generator being configured to generate a control pulse corresponding to a hold time of the first buffer stage such that the buffer circuit detects a transition of the input signal during the hold time.

Abstract: A method and apparatus are provided for error correction of a memory by using a first memory (18) and second memory (14) to perform error correction code (ECC) processing on data retrieved from the first memory and to use status control bits (35-37) in the second memory to detect and manage hard and soft errors identified by the ECC processing.