Abstract: A system is provided for transmitting signals. The system includes a ground-referenced single-ended signaling (GRS) driver circuit that is configured to pre-charge a first capacitor to store a first charge between a first output node and a first reference node based on a first input data signal during a first pre-charge phase and drive an output signal relative to a ground network based on the first charge during a first drive phase. A control circuit is configured to generate a first set of control signals based on the first input data signal and a first clock signal, where the first set of control signals causes the first GRS driver circuit to operate in either the first pre-charge phase or in the first drive phase.

Abstract: A system is provided for transmitting signals. The system comprises a first processing unit, a memory subsystem, and a package. The first processing unit is configured to include a first ground-referenced single-ended signaling (GRS) interface circuit. The memory subsystem is configured to include a second GRS interface circuit. The package is configured to include one or more electrical traces that couple the first GRS interface to the second GRS interface, where the first GRS interface circuit and the second GRS interface circuit are each configured to transmit a pulse along one trace of the one or more electrical traces by discharging a capacitor between the one trace and a ground network.

Abstract: An equalizer provided in a digital transmitter compensates for attenuation in a signal channel to a digital receiver. The equalizer generates signal levels as a logical function of bit history to emphasize transition signal levels relative to repeated signal levels. The preferred equalizer includes an FIR transition filter using a look-up table. Parallel circuits including FIR filters and digital-to-analog converters provide a high speed equalizer with lower speed circuitry. The equalizer is particularly suited to in-cabinet and local area network transmissions where feedback circuitry facilitates adaptive training of the equalizer.

Abstract: Systems and methods for managing context switches among threads in a processing system. A processor may perform a context switch between threads using separate context registers. A context switch allows a processor to switch from processing a thread that is waiting for data to one that is ready for additional processing. The processor includes control registers with entries which may indicate that an associated context is waiting for data from an external source.

Abstract: A method and a system are provided for synchronizing a signal. A keep out window is defined relative to a second clock signal and an edge detection signal is generated that indicates if an edge of a first clock signal is within the keep out window. The edge detection signal may be filtered. An input signal is received in a domain corresponding to the first clock signal and a delayed input signal is generated. Based on the edge detection signal or the filtered edge detection signal, either the input signal or the delayed input signal is selected, to produce an output signal in a domain corresponding to the second clock signal.

Abstract: A system, method, and computer program product are provided for compiling a computer program comprising arithmetic operations having different requirements with respect to numeric dynamic range, numeric resolution, or any combination thereof. The method comprises generating a transformed graph representation of the computer program by applying propagation rules that provide for relaxed numeric requirements, where applicable, and generating output code based on the transformed graph representation. Relaxing numeric requirements, such as dynamic range and resolution requirements, may advantageously lower power consumption during execution of the computer program.

Abstract: A system, method, and computer program product are provided for executing casting-arithmetic instructions. The method comprises receiving a casting-arithmetic instruction that specifies an arithmetic operation to be performed on input data and at least one casting operation of an input casting operation and an output casting operation. Upon determining that the casting-arithmetic instruction specifies the input casting operation, the input casting operation is performed on identified terms comprising the input data. Then the arithmetic operation is performed on the input data to generate an arithmetic result. Upon determining that the casting-arithmetic instruction specifies the output casting operation, the output casting operation is performed on the arithmetic result.

Abstract: A system includes a control circuit and first, second, and third ground-referenced single-ended signaling (GRS) driver circuits that are each coupled to an output signal. The control circuit is configured to generate a first, second, and third set of control signals that are each based on a respective phase of a clock signal. Each GRS driver circuit is configured to pre-charge a capacitor to store a charge based on the respective set of control signals during at least one phase of the clock signal and drive the output signal relative to a ground network by discharging the charge during a respective phase of the clock signal.

Abstract: One embodiment of the present invention sets forth a technique for capturing and holding a level of an input signal using a latch circuit that presents a low number of loads to the clock signal. The clock is only coupled to a bridging transistor and a pair of clock-activated pull-down or pull-up transistors. The level of the input signal is propagated to the output signal when the storage sub-circuit is not enabled. The storage sub-circuit is enabled by the bridging transistor and a propagation sub-circuit is activated and deactivated by the pair of clock-activated transistors.

Abstract: A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes, a plurality of routers, each router directly coupled to a plurality of terminal nodes, the routers coupled to one another and arranged into a group, and a plurality of groups of routers, such that each group is connected to each other group via at least one direct connection.

Abstract: One embodiment of the present invention sets forth a mechanism for transmitting and receiving ground-referenced single-ended signals. A transmitter combines a direct current (DC) to DC converter including a flying capacitor with a 2:1 clocked multiplexer to drive a single-ended signaling line. The transmitter drives a pair of voltages that are symmetric about the ground power supply level. Signaling currents are returned to the ground plane to minimize the generation of noise that is a source of crosstalk between different signaling lines. Noise introduced through the power supply is correlated with the switching rate of the data and may be reduced using an equalizer circuit.

Abstract: A system and method are provided for configuring a plurality of pin resources. The method includes identifying a plurality of pin resources of a primary application specific integrated circuit (ASIC) device and configuring the plurality of pin resources based on a pin distribution between a first interface and a second interface, where the first interface provides a first communication path between the primary ASIC device and a first device, and the second interface provides a second communication path between the primary ASIC device and a second device.

Abstract: Systems and methods for managing context switches among threads in a processing system. A processor may perform a context switch between threads using separate context registers. A context switch allows a processor to switch from processing a thread that is waiting for data to one that is ready for additional processing. The processor includes control registers with entries which may indicate that an associated context is waiting for data from an external source.

Abstract: Aspects relate to methods, devices and manufacturing relating to routing networks including a method for routing data units. The data units are individually routable through a routing network. A reservation request data unit is received over the routing network and from a sender. At least one speculative data unit associated with the reservation request data unit from the sender is received. The at least one speculative data unit is dropped in response to the at least one speculative data unit being blocked within the routing network. The sender is provided, over the routing network, a negative acknowledgement data unit that indicates the dropping of the at least one speculative data unit. The sender is provided a grant data unit indicating a start time. After the start time, at least one non-speculative data unit corresponding to the reservation request from the sender is received.

Abstract: A system and method are provided for controlling a switching voltage regulator circuit. An energy difference between a stored energy of a switching voltage regulator and a target energy is determined. A control variable of the switching voltage regulator is computed based on the energy difference and the control variable is applied to a current control mechanism of the switching voltage regulator. In one embodiment, the control variable is pulse width of a control signal.

Abstract: An equalizer provided in a digital transmitter compensates for attenuation in a signal channel to a digital receiver. The equalizer generates signal levels as a logical function of bit history to emphasize transition signal levels relative to repeated signal levels. The preferred equalizer includes an FIR transition filter using a look-up table. Parallel circuits including FIR filters and digital-to-analog converters provide a high speed equalizer with lower speed circuitry. The equalizer is particularly suited to in-cabinet and local area network transmissions where feedback circuitry facilitates adaptive training of the equalizer.

Abstract: Atomic memory access requests are handled using a variety of systems and methods. According to one example method, a data-processing circuit having an address-request generator that issues requests to a common memory implements a method of processing the requests using a memory-access intervention circuit coupled between the generator and the common memory. The method identifies a current atomic-memory access request from a plurality of memory access requests. A data set is stored that corresponds to the current atomic-memory access request in a data storage circuit within the intervention circuit. It is determined whether the current atomic-memory access request corresponds to at least one previously-stored atomic-memory access request. In response to determining correspondence, the current request is implemented by retrieving data from the common memory. The data is modified in response to the current request and at least one other access request in the memory-access intervention circuit.

Abstract: A method and a system are provided for synchronizing a signal. A keep out window is defined relative to a second clock signal and an edge detection signal is generated that indicates if an edge of a first clock signal is within the keep out window. The edge detection signal may be filtered. An input signal is received in a domain corresponding to the first clock signal and a delayed input signal is generated. Based on the edge detection signal or the filtered edge detection signal, either the input signal or the delayed input signal is selected, to produce an output signal in a domain corresponding to the second clock signal.

Abstract: A system, method, and computer program product for converting a design from edge-triggered docking to two-phase non-overlapping clocking is disclosed. The method includes the steps of replacing an edge-triggered flip-flop circuit that is coupled to a combinational logic circuit with a pair of latches including a first latch circuit and a second latch circuit and determining a midpoint of the combinational logic circuit based on timing information. The second latch circuit is propagated to a midpoint of the combinational logic circuit and two-phase non-overlapping clock signals are provided to the pair of latches.

Abstract: An equalizer provided in a digital transmitter compensates for attenuation in a signal channel to a digital receiver. The equalizer generates signal levels as a logical function of bit history to emphasize transition signal levels relative to repeated signal levels. The preferred equalizer includes an FIR transition filter using a look-up table. Parallel circuits including FIR filters and digital-to-analog converters provide a high speed equalizer with lower speed circuitry. The equalizer is particularly suited to in-cabinet and local area network transmissions where feedback circuitry facilitates adaptive training of the equalizer.