Abstract:
A system for parallelizing software in computer-aided design (CAD) software for circuit design includes a computer. The computer is configured to form or optimize a plurality of clusters in parallel. Each cluster in the plurality of clusters includes a set of nodes in a netlist in a design. The computer is configured to determine placements for blocks in a netlist in parallel, based on iterative improvement, partitioning, or analytic techniques.
Abstract:
A method is provided for mapping logic design memory into physical memory devices of a programmable logic device. User constraints and physical constraints may be taken into account in generating the mapping solution. Functional block layout on the programmable logic device may be taken into account when generating the mapping solution. Multiple types of physical memory types may be considered and logic design memory may be mapped to those types of physical memory devices that are determined to be the most appropriate. A mapping solution may be optimized using, for example, simulated annealing.
Abstract:
A method is provided for mapping logic design memory into physical memory devices of a programmable logic device. User constraints and physical constraints may be taken into account in generating the mapping solution. Functional block layout on the programmable logic device may be taken into account when generating the mapping solution. Multiple types of physical memory types may be considered and logic design memory may be mapped to those types of physical memory devices that are determined to be the most appropriate. A mapping solution may be optimized using, for example, simulated annealing.
Abstract:
A method for designing a system on a target device includes identifying candidate portions in the system to preserve based on similarities between the system and another system. Preservation criteria are applied on the candidate portions in the system to preserve to identify portions of the system to preserve. Design results from the another system are reused for portions in the system that are preserved.
Abstract:
A method is provided for mapping logic design memory into physical memory devices of a programmable logic device. User constraints and physical constraints may be taken into account in generating the mapping solution. Functional block layout on the programmable logic device may be taken into account when generating the mapping solution. Multiple types of physical memory types may be considered and logic design memory may be mapped to those types of physical memory devices that are determined to be the most appropriate. A mapping solution may be optimized using, for example, simulated annealing.
Abstract:
A method for designing a system on a target device includes identifying candidate portions in the system to preserve based on similarities between the system and another system. Preservation criteria are applied on the candidate portions in the system to preserve to identify portions of the system to preserve. Design results from the another system are reused for portions in the system that are preserved.
Abstract:
Systems and techniques are described for optimizing placement and routing by providing global information during early stages of a computer aided design (CAD) flow to produce better place and route solutions. Moreover, the systems and techniques described herein use natural connectivity information inherently provided in a design hierarchy.
Abstract:
A system for providing parallelization in computer aided design (CAD) software includes a computer. The computer is configured to identify a set of tasks having local independence, and assign each task in the set of tasks to be performed in parallel. The computer is further configured to perform each task in the set of tasks.
Abstract:
A method for performing routing for a logic design includes utilizing signal transition time as a criteria for selecting resources to provide interconnection between the components of the logic design.
Abstract:
A circuit with N primary outputs and a delay chain with M selection multiplexers. M can be less than N, and M is based on the number of primary outputs that simultaneously require a delayed signal from the delay chain. The N primary outputs may include core outputs and/or registers. Each of the M selection multiplexers feed directly or indirectly a subset of the N primary outputs.