Sciweavers

DAC
2008
ACM

Enhancing timing-driven FPGA placement for pipelined netlists

14 years 5 months ago
Enhancing timing-driven FPGA placement for pipelined netlists
FPGA application developers often attempt to use pipelining, Cslowing and retiming to improve the performance of their designs. Unfortunately, such registered netlists present a fundamentally different problem to the CAD tools, limiting the benefit of these techniques. In this paper we discuss some of the inherent quality and runtime issues pipelined netlists present to existing timingdriven placement approaches. We then present some algorithmic modifications that reduce post-compilation critical path delay by an average of 40%. Categories and Subject Descriptors B.7.2 [Integrated Circuits]: Design Aids ? Placement and routing. General Terms Algorithms, Design. Keywords Reconfigurable logic, FPGA, placement, simulated annealing, timing-driven, pipelined.
Kenneth Eguro, Scott Hauck
Added 12 Nov 2009
Updated 12 Nov 2009
Type Conference
Year 2008
Where DAC
Authors Kenneth Eguro, Scott Hauck
Comments (0)