Abstract: A method for implementing a convolutional neural network (CNN) accelerator on a target includes utilizing one or more processing elements to implement a standard convolution layer. A configuration of the CNN accelerator is modified to change a data flow between components on the CNN accelerator. The one or more processing elements is utilized to implement a fully connected layer in response to the change in the data flow.

Abstract: A method for designing a system on a target device includes generating a solution for the system. A solution for a module of the system identified by a user is preserved. The preserved solution for the module is implemented at a location on the target device identified by the user.

Abstract: A method for implementing a convolutional neural network (CNN) accelerator on a target includes identifying characteristics and parameters for the CNN accelerator. Resources on the target are identified. A design for the CNN accelerator is generated in response to the characteristics and parameters of the CNN accelerator and the resources on the target.

Abstract: A method for implementing a convolutional neural network (CNN) accelerator on a target includes utilizing one or more processing elements to implement a standard convolution layer. A configuration of the CNN accelerator is modified to change a data flow between components on the CNN accelerator. The one or more processing elements is utilized to implement a fully connected layer in response to the change in the data flow.

Abstract: A method for designing a system on a target device includes identifying a timing exception for a portion of a signal path. An area on the target device that includes components affected by the timing exception. Register retiming is performed where pipeline registers are added at boundaries of the area.

Abstract: A method for implementing a convolutional neural network (CNN) accelerator on a target includes utilizing one or more processing elements to implement a standard convolution layer. A configuration of the CNN accelerator is modified to change a data flow between components on the CNN accelerator. The one or more processing elements is utilized to implement a fully connected layer in response to the change in the data flow.

Abstract: A method for designing a system on a target device includes generating a solution for the system. A solution for a module of the system identified by a user is preserved. The preserved solution for the module is implemented at a location on the target device identified by the user.

Abstract: A method for designing a system on a target device includes generating a solution for the system. A solution for a module of the system identified by a user is preserved. The preserved solution for the module is implemented at a location on the target device identified by the user.

Abstract: A first circuit design description may have registers and combinational gates. Circuit design computing equipment may perform register retiming on the first circuit design description, whereby registers are moved across combinational gates during a first circuit design implementation. An engineering-change-order (ECO) of the first circuit design may result in a second circuit design. The differences between the first and second circuit designs may be confined to a region-of-change. The circuit design computing equipment may preserve the results from the first circuit design implementation and re-use portions of these results during the implementation of the second circuit design.

Abstract: A method for designing a system on a target device includes performing speculative register retiming with speculative changes made to a design of the system after an initial compilation of the design. A strategy is generated for an actual register retiming in response to user specified preferences on the speculative changes.

Abstract: A method for designing a system on a target device includes identifying a timing exception for a portion of a signal path. An area on the target device that includes components affected by the timing exception. Constraints are generated that prevent registers residing in the area from being used for register retiming.

Abstract: An integrated circuit may include path delay calibration circuitry. The calibration circuitry may be configured to calibrate respective delay paths so that data and control signals travelling through the respective delay paths experience proper propagation delays during normal user operation. The calibration circuitry may include a high frequency error calibration circuit, a monitoring circuit, and a calibration processing circuit. The high frequency error calibration circuit may be used to compute first calibration settings that take into account jitter and process variations. The monitoring circuit may be used to measure a proxy parameter of interest. The processing circuit may be used to compute an offset based at least partly on the measured value of the proxy parameter. The offset may be applied to the first calibration settings to obtain second calibration settings, which can be used to configure the respective delay paths.

Abstract: A method for designing a system on a target device includes generating a solution for the system. A solution for a module of the system identified by a user is preserved. The preserved solution for the module is implemented at a location on the target device identified by the user.

Abstract: In one embodiment, a tangible, non-transitory, computer-readable medium, includes instructions to receive a first circuit design, determine one or more variations of the first circuit design using register retiming with speculative circuit design changes, determine one or more performance improvements of the variations when fed clock signals over the first circuit design, determine one or more tradeoffs of the one or more variations of the first circuit design in comparison to the first circuit design, display a summary of the one or more variations of the first circuit design, the one or more performance improvements, and the one or more tradeoffs, and provide a user-selectable user interface element to enable selection of the first circuit design, at least one of the one or more variations of the first circuit design, or a combination thereof.

Abstract: Circuit design computing equipment may perform register retiming operations to improve the performance of a circuit design after having performed placement and routing operations. For example, the circuit design computing equipment may perform register retiming operations that move registers from a first portion of a circuit design that operates in a first clock domain into a synchronization region that separates the first portion of the circuit design from a second portion of the circuit design that operates in a second clock domain that is different than the first clock domain. Performing register retiming operations that move registers into a synchronization region between clock domains may solve the so-called short path—long path problem in which a long path that would benefit from a register retiming operation is coupled in parallel to a short path that has no location to receive a register during the register retiming operation.

Abstract: A method for designing a system on a target device includes generating a solution for the system. A solution for a module of the system identified by a user is preserved. The preserved solution for the module is implemented at a location on the target device identified by the user.

Abstract: A first circuit design description may have registers and combinational gates. Circuit design computing equipment may perform register retiming on the first circuit design description, whereby registers are moved across combinational gates during a first circuit design implementation. An engineering-change-order (ECO) of the first circuit design may result in a second circuit design. The differences between the first and second circuit designs may be confined to a region-of-change. The circuit design computing equipment may preserve the results from the first circuit design implementation and re-use portions of these results during the implementation of the second circuit design.

Abstract: Devices and methods for filtering data include calculating intermediate input values from input elements using a transformation function. The transformation function is based at least in part on a size of the filter and a number of filter outputs. Intermediate filter values are calculated from filter elements of the filter using the transformation function. Each intermediate input value is multiplied with a respective intermediate filter value to form intermediate values. These intermediate values are combined with each other using the transformation function to determine one or more output values.

Abstract: A method for designing a system on a target device includes identifying a timing exception for a portion of a signal path. An area on the target device that includes components affected by the timing exception. Register retiming is performed where pipeline registers are added at boundaries of the area.

Abstract: A first circuit design description may have registers and combinational gates. Circuit design computing equipment may perform register retiming on the first circuit design description, whereby registers are moved across combinational gates during a first circuit design implementation. An engineering-change-order (ECO) of the first circuit design may result in a second circuit design. The differences between the first and second circuit designs may be confined to a region-of-change. The circuit design computing equipment may preserve the results from the first circuit design implementation and re-use portions of these results during the implementation of the second circuit design.