Abstract: A compressed replay buffer in a first electronic unit of an electronic system holds commands in a table. As commands are transmitted from the first electronic unit to a second electronic unit, the command, along with associated data, command type, and the like are stored in a row in the table. No rows in the table contain “dead cycles” to indicate that no command was sent on a particular cycle on a bus over which the commands were transmitted. The second electronic unit may request that the first electronic unit replay some number of commands. In response, the first electronic unit uses commands in the compressed replay buffer, along with required timings stored on the first electronic unit, to replay the number of commands requested.

Abstract: An electronic system having a spread spectrum clock is disclosed. A spread spectrum clock source creates and transmits both a spread spectrum clock signal and a modulation signal. A spread spectrum clock generator uses a modulation waveform on the modulation signal to frequency modulate a reference oscillator frequency. A logic unit comprises a Phase Locked Loop that receives the spread spectrum clock signal and the modulation signal and generates a logic unit clock signal.

Abstract: A static random access memory (SRAM) includes an SRAM cell to store a bit of data. A word line accesses the SRAM cell, which, responsively, during a read, drives either a bit line true (BLT) or a bit line complement (BLC) low. Both BLT and BLC are precharged to a supply voltage, then, subsequently are discharged to a reference voltage, lower than the supply voltage, prior to the word line being activated. Because the bit lines are at a voltage lower than the supply voltage when the SRAM cell is activated, the SRAM cell stability is improved.

Abstract: Vertically stacked Field Effect Transistors (FETs) are created where a first FET and a second FET are controllable independently. The vertically stacked FETs may be connected in series or in parallel, thereby suitable for use as a portion of a NAND circuit or a NOR circuit. Epitaxial growth over a source and drain of a first FET, and having similar doping to the source and drain of the first FET provide a source and drain of a second FET. An additional epitaxial growth of a type opposite the doping of the source and drain of the first FET provides a body for the second FET.

Abstract: A method and apparatus for determining correct timing for receiving, in a host in a memory system, of a normal toggle transmitted by an addressed memory chip on a bidirectional data strobe. An offset in the data strobe is established, either by commanding the addressed memory chip, in a training period, to drive the data strobe to a known state, except during transmission of a normal toggle, or by providing a voltage offset between a true and a complement phase in the data strobe, or by providing a circuit bias in a differential receiver on the host the receives the data strobe. A series of read commands are transmitted by the host to the addressed memory chip, which responds by transmitting the normal toggle. Timing of reception of the normal toggle as received by the host chip is adjusted until the normal toggle is correctly received.

Abstract: A logic arrangement and method to support implied storage operation decode uses redundant target address detection, whereby target addresses of previous instructions are compared with the target address of the current instruction, and if equal, and the target addresses of previous instructions are not used as sources, the current instruction is decoded as a store instruction. This allows a redundant operation in an instruction set architecture to be redefined as a store instruction, freeing up opcodes normally used for store instructions to be used for other instructions.

Abstract: A particular chip is designed having a first variant (front side connected chip) of the chip and a second variant (back side connected chip). The first variant of the chip is attached to a carrier. The second variant of the chip is attached to the carrier inverted relative to the first variant of the chip. The first and second variants of the chip are attached to the carrier such that a vertical surface (side) of the first variant of the chip faces a corresponding vertical surface of the second variant of the chip. A circuit on the first variant of the chip is electrically connected to a corresponding circuit on the second variant of the chip.

Abstract: The present invention provides a method and apparatus for partitioning, sorting a data set on a multi-processor system. Herein, the multi-processor system has at least one core processor and a plurality of accelerators.

Abstract: A method and apparatus to improve memory initialization in a memory of a computer system. Memory units in the memory comprise a plurality of ranks, each rank having a unique rank select. A parity generator outputs a parity bit corresponding to whether an encoded rank select has an even or odd number of “1”s. The parity bit is used by an Error Checking and Correcting (ECC) unit that generates ECC bits that are stored in a rank having an active rank select. During a first interval in a memory initialization period, ranks having an even number of “1”s in their encoded rank select are initialized in parallel. During a second interval in the memory initialization period, ranks having an odd number of “1”s in their encoded rank select are initialized in parallel.

Abstract: An enhanced FET capable of controlling current above and below a gate of the FET. The FET is formed on a semiconductor substrate. A source and drain are formed in the substrate (or in a well in the substrate). A first epitaxial layer of similar doping to the source and drain are grown on the source and drain, the first epitaxial layer is thicker than the gate, but not so thick as to cover the top of the gate. A second epitaxial layer of opposite doping is grown on the first epitaxial layer thick enough to cover the top of the gate. The portion of the second epitaxial layer above the gate serves as a body through which the gate controls current flow between portions of the first epitaxial layer over the drain and the source.

Abstract: A 3-D (Three Dimensional) inverter having a single gate electrode. The single gate electrode has a first gate dielectric between the gate electrode and a body of a first FET (Field Effect transistor) of a first doping type, the first FET having first source/drain regions in a semiconductor substrate, or in a well in the semiconductor substrate. The single gate electrode has a second gate dielectric between the gate electrode and a body of a second FET of opposite doping to the first FET. Second source/drain regions of the second FET are formed from epitaxial layers grown over the first source/drain regions.

Abstract: To ensure efficient access to a memory whose writing process is slow. There is provided a storage device for caching data read from a main memory and data to be written in the main memory, comprises a cache memory having a plurality of cache segments, one or more cache segments holding data matching with data in the main memory being set in a protected state to protect the cache segments from a rewrite state, an upper limit of a number of the one or more cache segments being a predetermined reference number; and a cache controller that, in accordance with a write cache miss, allocates a cache segment selected from those cache segments which are not in the protected state to cache write data and writes the write data in the selected cache segment.

Abstract: A method is provided that utilizes silicon through via technology, to build a Toroid into the chip with the addition of a layer of magnetic material such as Nickel above and below the T-coil stacked multi-ring structure. This allows the connection between the inner via and an array of outer vias. This material is added on a BEOL metal layer or as an external coating on the finished silicon. Depending on the configuration and material used for the via, the inductance will increase approximately two orders of magnitude (e.g., by utilizing a nickel via core). Moreover, a ferrite material with proper thermal conduction properties is used in one embodiment.

Abstract: An apparatus and method are disclosed for an improved semiconductor interconnect scheme using a simplified process. In an embodiment of the apparatus, a polysilicon shape is formed on a silicon area. The polysilicon shape is created having a bridging vertex. When a spacer is created on the polysilicon shape, the spacer width is formed to be small enough near the bridging vertex to allow a silicide bridge to form that creates an electrical coupling between the silicon area and the bridging vertex. Semiconductor devices and circuits are created using the improved semiconductor interconnect scheme using the simplified process.

Abstract: The present invention provides a refresh controller for embedded DRAM, configured to receive an external access signal and generate refresh enabling signal REFN, refresh address signal CRA and confliction signal, said embedded DRAM comprising a plurality of memory groups, said controller comprising: a status controlling module that generates refresh enabling signal REFN and last refresh signal last_ccr according to the refresh interval and clock cycles; a refresh searching module that searches in said plurality of memory bank groups for at least one memory bank group that is to be refreshed in the refresh interval, and generates refresh address signal CRA according to the external access signal and the searched memory bank group; a scoreboard module that records the status of each of said plurality of memory bank groups according to said refresh address signal CRA and external access signal; and a confliction detecting module that generates confliction signal according to said external access signal, last ref

Abstract: A method and apparatus to efficiently scrub a memory, during a scrub period, of a computer system that has a memory comprising a number of memory elements. Examples of memory elements are memory ranks and banks. A memory rank may further comprise one or more banks. The computer system has a memory controller that receives read requests and write requests from a processor. The memory controller includes a scrub controller configured to output more than one scrub request during a particular request selector cycle. The memory controller includes a request selector that services a read request, a write request, or one of the scrub requests during a request selector cycle.

Abstract: A method and apparatus to scrub a memory during a scrub period, of a computer system. The computer system has a memory controller that receives read requests and write requests from a processor. The memory controller provides a different priority for scrub requests versus read requests during a period of relatively light memory workload versus a period of relatively heavy workload. The memory controller provides a relatively higher priority for scrub requests near an end of a scrub period if scrub progress is behind an expected scrub progress.

Abstract: A 3-D (Three Dimensional) inverter having a single gate electrode. The single gate electrode has a first gate dielectric between the gate electrode and a body of a first FET (Field Effect transistor) of a first doping type, the first FET having first source/drain regions in a semiconductor substrate, or in a well in the semiconductor substrate. The single gate electrode has a second gate dielectric between the gate electrode and a body of a second FET of opposite doping to the first FET. Second source/drain regions of the second FET are formed from epitaxial layers grown over the first source/drain regions.

Abstract: An apparatus and method is disclosed for providing capacity on demand using control to alter latency and/or bandwidth on a signaling bus in a computer system. If additional capacity is required, authorization is requested for additional capacity. If authorized, bandwidth of the signaling bus is increased to provide additional capacity in the computing system. Alternatively, upon authorization, latency of data transmissions over the signaling bus is reduced. In another alternative, upon authorization, memory timings are adjusted to speed up memory fetches and stores.

Abstract: A memory system having a data bus coupling a memory controller and a memory. The data bus has a number of data bus bits. The data bus is programmably apportioned to a first portion dedicated to transmitting data from the memory controller to the memory and a second portion dedicated to transmitting data from the memory to the memory controller. The apportionment can be assigned by suitable connection of pins on a memory chip in the memory and the memory controller to logical values. Alternatively, the apportionment can be scanned into the memory controller and the memory at bring up time. In another alternative, the apportionment can be changed by suspending data transfer and dynamically changing the sizes of the first portion and the second portion.