摘要:
A circuit measures the signal propagation delay through a selected test circuit. The test circuit is provided with an inverting feedback path so that the test circuit and feedback path together form a free-running oscillator. The oscillator then automatically provides its own test signal that includes alternating rising and falling signal transitions on the test-circuit input node. These signal transitions are counted over a predetermined time period to establish the average period of the oscillator. Finally, the average period of the oscillator is related to the average signal propagation delay through the test circuit. One embodiment of the invention includes a phase discriminator that samples the output of the oscillator and accumulates data representing the duty cycle of that signal. The duty cycle can then be combined with the average period of the test signal to determine, separately, the delays associated with falling and rising edges propagating through the test circuit.
摘要:
A circuit measures the signal propagation delay through a selected test circuit. The test circuit is provided with a feedback path so that the test circuit and feedback path together form a free-running oscillator. The oscillator then automatically provides its own test signal that includes alternating rising and falling signal transitions on the test-circuit input node. These signal transitions are counted over a predetermined time period to establish the average period of the oscillator. Finally, the average period of the oscillator is related to the average signal propagation delay through the test circuit. A phase discriminator samples the output of the oscillator and accumulates data representing the duty cycle of that signal. The duty cycle can then be combined with the average period of the test signal to determine, separately, the delays associated with falling and rising edges propagating through the test circuit.
摘要:
A circuit measures a signal propagation delay through a series of memory elements. In one embodiment the memory elements are configured in series so that together they form a delay circuit. In another embodiment the memory elements are configured in a loop to form a ring oscillator. Each memory element propagates a signal to a subsequent memory element so that the time the signal takes to traverse all of the memory elements is proportional to the average delay induced by the individual elements. This proportionality provides an effective means for measuring the delays of those components. Various embodiments of the invention measure the speeds at which memory elements can be preset, cleared, written to, read from, or clock enabled.
摘要:
A circuit separately measures a selected one of the rising-edge and falling-edge signal propagation delays through one or more circuits of interest. A number of synchronous components are configured in a loop so that they together form a free-running ring oscillator. Each synchronous component clocks a subsequent synchronous component in the ring; the subsequent synchronous component responds by clocking a later component in the ring and by clearing a previous component to prepare it for a subsequent clock. The oscillator thus produces an oscillating test signal in which the period is proportional to the clock-to-out delays of synchronous components. This proportionality provides an effective means for measuring the clock-to-out delays of those components. Other embodiments include additional asynchronous test circuit paths for which the associated signal propagation delays are of interest.
摘要:
A tunable clock distribution system is used to minimize the power dissipation of a clock distribution network in an integrated circuit. The tunable clock distribution system provides a tunable inductance on the clock distribution network to adjust a resonant frequency in the tunable clock distribution system. The inductance is tuned so that the resonant frequency of the tunable clock distribution system approaches the frequency of the clock signal on the clock distribution network. As the resonant frequency of the tunable clock distribution system approaches the frequency of the clock signal, the power dissipation of the clock distribution network decreases. Some embodiments also provide a tunable capacitance on the clock distribution network to adjust the resonant frequency of the tunable clock distribution system.
摘要:
Methods of enabling functions of a design to be implemented in an integrated circuit device are disclosed. An exemplary method comprises applying test data to a plurality of dice having different element types for implementing circuits, wherein the plurality of dice have a common layout of the different element types for implementing the circuits; receiving output data from the plurality of dice in response to applying the test data to the plurality of dice; analyzing the output data from the plurality of dice; transforming by a computer the output data to characterization data comprising timing data associated with the different element types for implementing circuits, wherein the characterization data comprises data associated with regions of the dice, and storing the characterization data. A computer program product for enabling functions of a design to be implemented in an integrated circuit device is also disclosed.
摘要:
A circuit measures a signal propagation delay through a series of memory cells on a programmable logic device. In one embodiment, a number of RAM cells are configured in series. Each RAM cell is initialized to store a logic zero. The first RAM cell is then clocked so that the output of the RAM cell rises to a logic one. The resulting rising edge from the output of the RAM cell then clocks the second RAM cell, which in turn clocks the next RAM cell in the series. The time required for a rising edge to traverse the entire sequence of latches is the cumulative time required for the output of each RAM cell to change in response to a clock edge. Consequently, the delay through the series of RAM cells provides a measure of the time required for one of the RAM cells to store data in response to a clock edge. In another embodiment, the RAM cells are arranged in a loop so that the sequence of RAM cells forms a ring oscillator, the period of which provides an indication of the time required for the RAM cells to store data in response to clock edges.
摘要:
A method of automatically optimizing a hierarchical netlist of integrated circuit cells comprising at least one upper-level cell containing a multiplicity of subsidiary cells of lower hierachical level includes receiving data defining said netlist and timing constraints for it, and establishing abstract timing models for all the subsidiary cells. Timing constraints are propagated to at least one selected subsidiary cell and this cell is optimized by means of a flat optimizer to produced an optimized version of the selected subsidiary cell. The optimized version of the selected cell is inserted into the netlist. The timing constraints denote arrival times for signals at inputs of a cell and required times for signals at outputs of a cell and each abstract timing model of a cell comprises timing parameters which enable a delay time between a specified input of a cell to a specified output of a cell to be computed.
摘要:
A method and apparatus is disclosed for detecting edge-sensitive behavior from HDL descriptions of a circuit and inferring a hardware implementation of that behavior as a generalized edge-triggered D-type flip-flop with asynchronous set and clear inputs. The invention detects the edge-sensitive behavior from directed acyclic graphs (DAGS) that represent the individual signal nets of the circuit as affected by each process defined in the HDL description of the circuit. The invention then modifies each DAG to infer the asychronous control expressions and the data input expression necessary to control generalized flip-flop to emulate the behavior of the net represented by the DAG. The invention then creates a symbolic hardware implementation of the net's behavior using the D-type flip-flop and any combinational logic necessary to produced the inferred control signals. The symbolic hardware implementations for each net can then be optimized using well-known techniques, and a netlist generated therefrom for purposes of creating masks for manufacturing the circuit. The invention can be easily implemented within known symbolic simulator routines already capable of synthesizing level-sensitive behavior using combinational logic.
摘要:
One or more embodiments provide a load balancing solution for improving the runtime performance of parallel HDL simulators. During compilation each process is analyzed to determine a simulation cost based on complexity of the HDL processes. During simulation, processes to be executed in the same simulation cycle are scheduled using the simulation costs computed at compile-time in order to reduce the delay incurred during simulation.