Abstract: An electronic system includes a mapping circuit configured to receive input samples of a dataset within a defined range of values. The mapping circuit is configured to perform comparisons that compare each input sample to each of a plurality of comparison values selected from the defined range of values. For each comparison, the mapping circuit generates an indication value specifying whether the input sample used in the comparison is greater than or equal to the comparison value used in the comparison. The system includes an adder circuit configured to generate a sum of the indication values for each comparison value and a memory configured to maintain counts corresponding to the comparison values. The counts are updated by the respective sums. The system includes a threshold detection circuit configured to determine, for the dataset, a threshold value or threshold range based on the counts read from the memory.

Abstract: A programmable, non-linear (PNL) activation engine for a neural network is capable of receiving input data within a circuit. In response to receiving an instruction corresponding to the input data, the PNL activation engine is capable of selecting a first non-linear activation function from a plurality of non-linear activation functions by decoding the instruction. The PNL activation engine is capable of fetching a first set of coefficients corresponding to the first non-linear activation function from a memory. The PNL activation engine is capable of performing a polynomial approximation of the first non-linear activation function on the input data using the first set of coefficients. The PNL activation engine is capable of outputting a result from the polynomial approximation of the first non-linear activation function.

Abstract: An electronic system includes a mapping circuit configured to receive input samples of a dataset within a defined range of values. The mapping circuit is configured to perform comparisons that compare each input sample to each of a plurality of comparison values selected from the defined range of values. For each comparison, the mapping circuit generates an indication value specifying whether the input sample used in the comparison is greater than or equal to the comparison value used in the comparison. The system includes an adder circuit configured to generate a sum of the indication values for each comparison value and a memory configured to maintain counts corresponding to the comparison values. The counts are updated by the respective sums. The system includes a threshold detection circuit configured to determine, for the dataset, a threshold value or threshold range based on the counts read from the memory.

Abstract: Implementing a hardware description language (HDL) memory includes determining, using computer hardware, a width and a depth of the HDL memory specified as an HDL module for implementation in an integrated circuit (IC), partitioning, using the computer hardware, the HDL memory into a plurality of super slices corresponding to columns and the plurality of super slices into a plurality of super tiles arranged in rows. A heterogeneous memory array may be generated, using the computer hardware. The heterogeneous memory array is formed of different types of memory primitives of the IC. Input and output circuitry configured to access the heterogeneous memory array can be generated using the computer hardware.

Abstract: Disclosed methods and systems involve, prior to mapping logic of the module to a target integrated circuit (IC) technology, estimating total delay of a module of a circuit design and determining whether or not the module is timing critical based on the total delay of the module and a timing constraint. Also prior to mapping, the module is restructured for timing optimization in response to determining that the module is timing critical. In response to determining that the module is not timing critical, and prior to mapping, the module is restructured for area optimization. The elements of the module are then mapped to the circuit elements of the target IC technology, followed by place-and-route and generating implementation data for making an IC that implements the circuit design.

Abstract: Determining on-chip memory access patterns can include modifying a circuit design to include a profiler circuit for a random-access memory (RAM) of the circuit design, wherein the profiler circuit is configured to monitor an address bus of the RAM, and modifying the circuit design to include a debug circuit connected to the profiler circuit. Usage data for the RAM can be generated by detecting, using the profiler circuit, addresses of the RAM accessed during a test of the circuit design, as implemented in an integrated circuit. The usage data for the RAM can be output using the debug circuit.

Abstract: Embodiments herein describe techniques for designing a compressed hardware implementation of a user-designed memory. In one example, a user defines a memory in hardware description language (HDL) with a depth (D) and a width (W). To compress the memory, a synthesizer designs a core memory array representing the user-defined memory. Using addresses, the synthesizer can identify groups of nodes in the array that can be compressed into a memory element. The synthesizer designs input circuitry such as a data replicator and a write enable generator for generating the inputs and control signals for the groups. The synthesizer can then implement the design in an integrated circuit where each group of nodes maps to a single memory element, thereby resulting in a compressed design.

Abstract: Implementing a circuit design can include detecting, using computer hardware, a re-convergent section of a circuit design, masking, using the computer hardware, a sequential circuit element of the re-convergent section located between a start and an end of the re-convergent section, and performing, using the computer hardware, an optimization operation on combinatorial logic of the re-convergent section to create optimized combinatorial logic. Using the computer hardware, the optimized combinatorial logic of the re-convergent section can be mapped. Further, the re-convergent section can be modified subsequent to the mapping to match timing of the re-convergent section prior to the masking.

Abstract: A memory optimization method includes identifying, within a circuit design, a memory having an arithmetic operator at an output side and/or an input side of the memory. The memory may include a read-only memory (ROM). In some examples, an input of the arithmetic operator includes a constant value. In some embodiments, the memory optimization method further includes absorbing a function of the arithmetic operator into the memory. By way of example, the absorbing the function includes modifying contents of the memory based on the function of the arithmetic operator to provide an updated memory and removing the arithmetic operator from the circuit design.

Abstract: Local retiming for a circuit design includes determining, using computer hardware, a load of a synchronous circuit element within the circuit design tagged for forward retiming, traversing, using the computer hardware, each input of the load backward through the circuit design until a sequential circuit element or a primary input is reached, and adding, using the computer hardware, each synchronous circuit element encountered in the traversing to a forward retiming list. In response to determining that forward retiming criteria is met for the forward retiming list, the computer hardware modifies the circuit design by creating a new synchronous circuit element at an output of the load.

Abstract: Register pull-out for sequential circuit blocks may include determining, using computer hardware, a net of a circuit design having a driver that is a macro circuit block driving a plurality of loads and determining, using the computer hardware, a placement difficulty of the net based upon a type of the driver and number and type of the plurality of loads. In response to determining that the placement difficulty of the net exceeds a threshold placement difficulty, the computer hardware is capable of modifying the circuit design by pulling a register from the driver to a location on a device external to the driver and changing internal logic of the driver based upon the pulled register.

Abstract: Verifying a circuit design may include, in response to modification of the circuit design involving at least one of inserting or removing a flip-flop, determining, using computer hardware, latency change values for pins of components of the circuit design, determining, using the computer hardware, total latency for the pins of the components of the circuit design based, at least in part, upon the latency change values, and comparing, using the computer hardware, total latency of the pins of the components of the circuit design. Verifying the circuit design may also include detecting, using the computer hardware, a latency error within the circuit design based upon the comparing and generating, using the computer hardware, a notification of the latency error in the circuit design, wherein the notification specifies a type of the latency error detected.

Abstract: Methods and apparatus relating generally to synthesis are described. In such a method, a directed graph for a circuit design is generated. A cascaded chain is identified in the directed graph with a timing violation. A pipeline register stage of the cascaded chain is moved (or added) to remove the timing violation. The circuit design is transformed to provide a netlist including the pipeline register stage.

Abstract: Reducing dynamic power consumption for a circuit can include analyzing, using a processor, a netlist specifying the circuit to determine a block of combinatorial circuitry in a first signal path with at least a threshold amount of switching activity and detecting, using the processor, a second signal path coupled to the block of combinatorial circuitry by a sequential circuit element. The second signal path has a delay that meets a target signal path requirement. Using the processor, the netlist can be modified by subdividing the block of combinatorial circuitry into at least a first portion and a second portion and moving one of the portions from the first signal path to the second signal path, wherein the moving separates the first portion from the second portion by the sequential circuit element.

Abstract: Disclosed approaches of processing a circuit design include determining a subset of addresses of a first RAM of the circuit design that are accessed more often than a frequency threshold. A specification of a second RAM is created for the subset of addresses. A decoder circuit is added to the circuit design. The decoder circuit is configured to enable the second RAM and disable the first RAM in response to an input address in the subset of addresses, and to enable the first RAM and disable the second RAM in response to an input address other than addresses in the subset of addresses.

Abstract: Reducing latency of a circuit design can include determining, using a processor, a set of sequential circuit elements of a circuit design that meets a condition for removal from the circuit design, wherein the condition is dependent upon a target technology process and a target operating frequency. Using the processor, a feasible cut for a selected sequential circuit element of the set is determined. The selected sequential circuit element and each other sequential circuit element of the set that is part of the cut is removed from the circuit design using the processor.

Abstract: Reducing dynamic power consumption for a circuit can include analyzing, using a processor, a netlist specifying the circuit to determine a block of combinatorial circuitry in a first signal path with at least a threshold amount of switching activity and detecting, using the processor, a second signal path coupled to the block of combinatorial circuitry by a sequential circuit element. The second signal path has a delay that meets a target signal path requirement. Using the processor, the netlist can be modified by subdividing the block of combinatorial circuitry into at least a first portion and a second portion and moving one of the portions from the first signal path to the second signal path, wherein the moving separates the first portion from the second portion by the sequential circuit element.