Abstract: A dictionary mapping source locale strings to target locale strings is constructed. A tree is constructed from the dictionary. The tree has nodes including a start node, end nodes, and string nodes. The nodes form node chains of the tree that correspond to the source locale strings. Each node chain links a collection of the nodes from the start node to a string node. Each node other than the start node has a parent node and corresponds to a character of the source locale strings. Each node other than the end nodes has one or more child nodes. Each string node ends a node chain, and corresponds to a mapping within the dictionary of a source locale string to a target locale string. An input string having the source locale is processed against the tree to generate an output string having the target locale.

Abstract: A method for designing integrated circuits uses clock signal interleaving to reduce the likelihood of a soft error arising from an upset in a clock distribution network. At least two circuits in a circuit description are identified as being sensitive to radiation, and different clock distribution nodes are assigned to the two circuits. Several exemplary implementations are disclosed. The second circuit may be a redundant replica of the first circuit, such as a reset circuit. The first and second circuits may be components of a modular redundant circuit such as a triple modular redundancy flip-flop. The first circuit may include a set of data bits for an entry of a storage array such as a register or memory array, and the second circuit may include a set of check bits associated with the entry.

Abstract: A multiplier circuit comprises a fused Booth encoder multiplexer which produces partial product bits, a tree which uses the partial product bits to generate partial products, and an adder which uses the partial products to generate intermediate sum and carry results for a multiplication operation. The fused Booth encoder multiplexer utilizes encoder-selector cells having a logic tree which carries out a Boolean function according to a Booth encoding and selection algorithm to produce one of the partial product bits at a dynamic node, and a latch connected to the dynamic node which maintains the value at an output node. The encoder-selector cells operate in parallel to produce the partial product bits generally simultaneously. A given one of the encoder-selector cells has a unique set of both multiplier operand inputs and multiplicand operand inputs, and produces a single partial product bit.

Abstract: A configurable PSRO measurement circuit is used to measure the frequency dependent capacitance of a target through silicon via (TSV) or other conductive structure. Measurements of the target structure are aided by using adjustable resistors and a de-embedding structure to measure the effects of parasitic capacitance, CPAR. Current is measured to both the device under test (DUT) and the de-embedding structure. From these measurements, the frequency dependent capacitance of the DUT is calculated.

Abstract: A distributed charge pump system uses a delay element and frequency dividers to generate out of phase pump clock signals that drive different charge pumps, to offset peak current clock edges for each charge pump and thereby reduce overall peak power. Clock signal division and phase offset may be extended to multiple levels for further smoothing of the pump clock signal transitions. A dual frequency divider may be used which receives the clock signal and its complement, and generates two divided signals that are 90° out of phase. In an illustrative embodiment the clock generator comprises a variable-frequency clock source, and a voltage regulator senses an output voltage of the charge pumps, generates a reference voltage based on a currently selected frequency of the variable-frequency clock source, and temporarily disables the charge pumps (by turning off local pump clocks) when the output voltage is greater than the reference voltage.

Abstract: Disclosed are a method, a system, and a program product for processing a data stream by accessing one or more hardware registers of a processor. In one or more embodiments, a first program instruction or subroutine can associate a hardware register of the processor with a data stream. With this association, the hardware register can be used as a stream head which can be used by multiple program instructions to access the data stream. In one or more embodiments, data from the data stream can be fetched automatically as needed and with one or more patterns which may include one or more start positions, one or more lengths, one or more strides, etc. to allow the cache to be populated with sufficient amounts of data to reduce memory latency and/or external memory bandwidth when executing an application which accesses the data stream through the one or more registers.

Abstract: Power, routability and electromigration have become crucial issues in modem microprocessor designs. In high performance designs, clocks are the highest consumer of power. Arranging clocking components with regularity so as to minimize the capacitance of the clock nets can help reduce clock power, however, it may hurt performance due to some loss of flexibility in physically placing those components. The present invention provides techniques to optimally design a clock network by logically assigning clusters of the latches to respective clock distribution structures, placing clock pins at favored pin locations, and placing clock distribution structures directly underneath the clock pins. The clock distribution structures may be moved to favored distribution locations along the clock stripes, and new optimal clustering generated between the latches and the clock distribution structures. These three optimizations are preferably repeated iteratively to derive a local optimal solution for the clock network.

Abstract: A contour of a mask design for an integrated circuit is modified to compensate for systematic variations arising from non-optical effects such as stress, well proximity, rapid thermal anneal, or spacer thickness. Electrical characteristics of a simulated integrated circuit chip fabricated using the mask design are extracted and compared to design specifications, and one or more edges of the contour are adjusted to reduce the systematic variation until the electrical characteristic is within specification. The particular electrical characteristic preferably depends on which layer is to be fabricated from the mask: on-current for a polysilicon; resistance for contact; resistance and capacitance for metal; current for active; and resistance for vias. For systematic threshold voltage variation, the contour is adjusted to match a gate length which corresponds to an on-current value according to pre-calculated curves for contour current and gate length at a nominal threshold voltage of the chip.

Abstract: A distributed charge pump system uses a delay element and frequency dividers to generate out of phase pump clock signals that drive different charge pumps, to offset peak current clock edges for each charge pump and thereby reduce overall peak power. Clock signal division and phase offset may be extended to multiple levels for further smoothing of the pump clock signal transitions. A dual frequency divider may be used which receives the clock signal and its complement, and generates two divided signals that are 90° out of phase. In an illustrative embodiment the clock generator comprises a variable-frequency clock source, and a voltage regulator senses an output voltage of the charge pumps, generates a reference voltage based on a currently selected frequency of the variable-frequency clock source, and temporarily disables the charge pumps (by turning off local pump clocks) when the output voltage is greater than the reference voltage.

Abstract: A timing-driven cloning method iteratively partitions sinks of the net into different sets of clusters and for each set computes a figure of merit for a cloned gate location which optimizes timing based on linear delay, that is, a delay proportional to the distance between the cloned gate location and the sinks. The set having the highest figure of merit is selected as the best solution. The original gate may also be moved to a timing-optimized location. The sinks are advantageously partitioned using boundaries of Voronoi polygons defined by a diamond region surrounding the original gate, or vice versa. The figure of merit may be for example worst slack, a sum of slacks at the sinks in the second cluster, or a linear combination of worst slack and sum of the slacks.

Abstract: A switched-capacitor charge pump comprises a two-phase charging circuit, cross-coupled transistors connected to output nodes of the switched capacitors, and a pump output connected to source terminals of the cross-coupled transistors. The charge pump has side transistors for boosting charge transfer, and gating logic of the side transistors includes level shifters which control connections to the pump output or a reference voltage. Negative and positive charge pump embodiments are provided. The charging circuit utilizes non-overlapping wide and narrow clock signals to generate multiple gating signals. The pump clock circuit preferably provides independent, programmable adjustment of the widths of the wide and narrow clock signals. An override mode can be provided using clamping circuits which shunt the pump output to the second nodes of the switched capacitors.

Abstract: A performance measure (e.g., processor speed) for computing components such as servers is optimized by creating models of power consumption versus the performance measure for each server, adding the power models to derive an overall power model, and calculating an optimum set point for the performance measure which corresponds to a power limit on the servers using the overall power model. The set point is then used to set power budgets for the servers based on their power models, and the servers maintain power levels no greater than their respective power budgets. The server power models are preferably created in real time by monitoring power consumption and the performance measure to derive sets of data points for the servers, and performing regression on the sets of data points to yield power models for the servers. Multiple server power models may be created for different program applications.

Abstract: Disclosed is a computer implemented method, computer program product, and apparatus for generating diagnostic data for a thread. A service processor may begin by executing the computer usable program code to allocate a first thread to a monitoring function. The service processor determines if a second thread is running, wherein the second thread is associated with an operation. The service processor obtains an object pointer from a list of running threads, wherein the object pointer references a time data structure associated with the second thread, the time data structure comprising a maximum running time interval and a check time. The service processor calculates the check time as a maximum running time interval added to an initially sampled timestamp. The service processor determines if a current time exceeds the check time. In addition, the service processor generates an error log, responsive to a determination that the current time exceeds the check time.

Abstract: This invention relates to a transparent and non-intrusive method for monitoring and managing the running of tasks executed in one or more computer processors, in particular in multi-processor systems with a parallel architecture. It proposes a system and method for managing a computer task, termed target, during a given execution period, termed activity period (SchJ, SchR), within a computer system, in a computer processor provided with means of monitoring or estimating performance and including a counter (PMC) with a given possible error in plus or minus, termed relative error, this process comprising on the one hand, an evaluation of a number of executed instructions (NR, NJ) up to at least one given point of said activity period, using said counter; and on the other hand, a generation of data, termed signature (SGJ, SGR), read or calculated from the state of the processor or computer system and corresponding to at least one given point of said activity period.