Abstract: A memory controller for heterogeneous computer processors dynamically adjusts access priorities by the different processors to maximize performance in the execution of a single parallel application program on both processor architectures. In one embodiment, the memory controller predicts sequential memory accesses by the processor having higher memory latency or fewer access requests to lockout the other processor during those sequences for improved implementation of the intended prioritization.

Abstract: A controller for voltage regulators providing power to computer processors may control the number of active phases of each voltage regulator according to a determined electrical current demand from the processor. By relying on electrical current demand rather than a P-state, the latter generally indicating a power conservation status, improved regulator efficiencies may be had, in particular responding to situations where low current demand occurs under heavy processor demand because of C-state variations.

Abstract: A fixed-point multiplier providing reduced energy usage dynamically truncates received operands according to the location of computationally important bits in the operands and provides the truncated operands to a reduced width multiplier offering reduced energy usage. Information about the location of the dynamic truncation is used to properly shift the result of the multiplier to provide an approximation of full multiplication of the operands.

Abstract: Sequential circuits with error-detection are provided. They may, for example, be used to replace traditional master-slave flip-flops, e.g., in critical path circuits to detect and initiate correction of late transitions at the input of the sequential. In some embodiments, such sequentials may comprise a transition detector with a time borrowing latch.

Abstract: Hardening of an integrated circuit such as a GPU processor to soft errors caused by particle strikes is applied selectively to the set of devices according to the magnitude of error resulting from this soft error for the particular device. This approach differs from approaches that protect all devices, all devices likely to produce an output error, or all devices that are vulnerable.

Abstract: A high-power-efficiency multiplier combines a standard floating-point multiplier with a power-of-two multiplier that performs multiplications by shifting operations without the need for floating-point multiplication circuitry. By selectively steering some operands to this power-of-two multiplier, substantial power savings may be realized. In one embodiment, multiplicands may be modified to work with the power-of-two multiplier introducing low errors that may be accommodated in pixel calculations.

Abstract: A graphic processing unit having multiple computational elements flexibly interconnected to memory elements provides for data compressors/decompressors in the memory channels communicating between the computational elements and memory elements to provide an effective increase in bandwidth of those connections by the compression of data transferred thereon.

Abstract: Sequential circuits with error-detection are provided. They may, for example, be used to replace traditional master-slave flip-flops, e.g., in critical path circuits to detect and initiate correction of late transitions at the input of the sequential. In some embodiments, such sequentials may comprise a transition detector with a time borrowing latch.

Abstract: A computer architecture provides both a parallel memory bus and serial memory bus between a processor system and memory. Latency-tolerant memory access requests are steered to the serial memory bus which operates to increase the available memory bus bandwidth on the parallel memory. The invention also provides integrated circuit computer memory suitable for this application.

Abstract: There is provided a susceptor. The susceptor includes: a body having a first surface, a second surface opposite the first surface, and an outer side surface connecting the first surface and the second surface; at least one pocket recessed from the first surface to accommodate at least one wafer therein, respectively; at least one tunnel respectively located below the pocket and extending from a center of the body to the outer side surface; at least one connecting channel each of which connects each of the pocket to each of the tunnel; and a supply line connected to the tunnel at the center of the body and supplying a gas from an outside in order for the gas to flow from the center of the body to the outer side surface.

Abstract: Hardening of an integrated circuit such as a GPU processor to soft errors caused by particle strikes is applied selectively to the set of devices according to the magnitude of error resulting from this soft error for the particular device. This approach differs from approaches that protect all devices, all devices likely to produce an output error, or all devices that are vulnerable.

Abstract: A memory structure is provided that controls the activation of error handling bits as a function of operating voltage. In this way, error correction can be used to offset errors when the memory structure is run at low voltage (and frequency). However, negative performance impacts for such error correction, such as additional access latencies, can be avoided when the memory structure is run at higher voltage (and frequency) and memory errors are less likely. In addition, increased latencies due to evaluating error handling bits may be hidden by reading digital data bits from the memory structures speculatively and assuming no errors. Also, certain portions of memory structures may have larger cells, and therefore larger areas, than other portions, which may provide not only higher reliability at low operating voltages, but also faster operation with reduced latency.

Abstract: A multicore computer architecture provides for clock dividers on each core, the clock dividers capable of providing rapid changes in the clock frequency of the core. The clock dividers are used to reduce the clock frequency of individual cores spinning while waiting for a synchronization instruction resolution such as a lock variable. Core power demands may be decreased before and after change in dock speed to reduce power bus disruption.

Abstract: A semiconductor light emitting device includes first conductivity type and second conductivity type semiconductor layers, an active layer disposed between the semiconductor layers and having a structure in which one or more quantum well layers and one or more quantum barrier layers are alternately disposed An electron blocking layer is disposed between the active layer and the second conductivity type semiconductor layer. A capping layer is disposed between the active layer and the electron blocking layer and blocking a dopant element from being injected into the active layer from the second conductivity type semiconductor layer.

Abstract: Aspects of the present invention relate to an approach for implanting and forming a polysilicon resistor with a single implant dose. Specifically, a mask having a set of openings is formed over a resistor surface. The set of openings are typically formed in a column-row arrangement according to a predetermined pattern. Forming the mask in this manner allows the resistor surface to have multiple regions/zones. A first region is defined by the set of openings in the mask, and a second region is defined by the remaining portions of the mask. The resistor is then subjected to a single implant dose via the openings. Implanting the resistor in this manner allows the resistor to have multiple resistance values (i.e., a first resistance value in the first region, and a second resistance value in the second region).

Abstract: A method includes providing a gate structure with at least one side wall and a bottom. At least one first spacer layer is formed over the at least one side wall. An offset spacer layer is formed over the at least one first spacer layer and the bottom. A bottom portion of the offset spacer layer is selectively removed to expose the bottom.

Abstract: A multi-core processor provides circuitry for jointly scaling the number of operating cores and the amount of resources per core in order to maximize processing performance in a power-constrained environment. Such scaling is advantageously provided without the need for scaling voltage and frequency. Selection of the number of operating cores and the amount of resources per core is made by examining the degree of instruction and thread level parallelism available for a given application. Accordingly, performance counters (and other characteristics) implemented in by a processor may be sampled on-line (in real time) and/or performance counters for a given application may be profiled and characterized off-line. As a result, improved processing performance may be achieved despite decreases in core operating voltages and increases in technology process variability over time.

Abstract: A method of forming a FinFET device involves performing an epitaxial growth process to form a layer of semiconducting material on a semiconducting substrate, wherein a first portion of the layer of semiconducting material will become a fin structure for the FinFET device and wherein a plurality of second portions of the layer of semiconducting material will become source/drain structures of the FinFET device, forming a gate insulation layer around at least a portion of the fin structure and forming a gate electrode above the gate insulation layer.

Abstract: A method for adjusting bandwidth, a bandwidth scaler and an apparatus are provided. The method for adjusting bandwidth involves determining a dynamic context of a processor, and based on the determined dynamic context, scaling bandwidth between the processor and a memory.

Abstract: An approach for controlling a critical dimension (CD) of a RMG of a semiconductor device is provided. Specifically, embodiments of the present invention allow for CD consistency between a dummy gate and a subsequent RMG. In a typical embodiment, a dummy gate having a cap layer is formed over a substrate. A re-oxide layer is then formed over the substrate and around the dummy gate. A set of doping implants will then be implanted in the substrate, and the re-oxide layer will subsequently be removed (after the set of doping implants have been implanted). A set of spacers will then be formed along a set of side walls of the dummy gate and an epitaxial layer will be formed around the set of side walls. Thereafter, the dummy gate will be replaced with a metal gate (e.g., an aluminum or tungsten body having a high-k metal liner there-around).