300 likes | 453 Views
ECpE 583 Reconfigurable Computing Lecture 22: Tue 11/11/2008 (Placing Applications onto FPGAs: Part III). Instructor: Dr. Phillip Jones (phjones@iastate.edu) Reconfigurable Computing Laboratory Iowa State University Ames, Iowa, USA http://www.ece.iastate.edu
E N D
ECpE 583Reconfigurable ComputingLecture 22: Tue 11/11/2008(Placing Applications onto FPGAs: Part III) Instructor: Dr. Phillip Jones (phjones@iastate.edu) Reconfigurable Computing Laboratory Iowa State University Ames, Iowa, USA http://www.ece.iastate.edu http://class.ece.iastate.edu/cpre583 (coming soon) http://www.arl.wustl.edu/~phjones/cpre583 (temporary)
Class Announcements • Updated schedule (coming soon) • MP2 concerns • CPRE594: Embedded System Optimization
Outline • Placing an application on an FPGA • Low-level details (Focus on Placement)
Applications on FPGA • Low-level: Application to FPGA refs • Reconfigurable Computing (2008) • Scott Hauck, Andre DeHon • The VLSI handbook (2000) • Wai-Kai Chen (note: there’s a 2nd edition (2006)) • Combinational Logic Synthesis for LUT Based Field Programmable Gate Arrays • Jason Cong (TODAES’96)
Applications on FPGA: Low-level • Implement circuit in VHDL (Verilog) • Simulate compiled VHDL • Synthesis VHDL into a device independent format • Map device independent format to device specific resources • Check that device has enough resources for the design • Place resources onto physical device locations • Route (connect) resources together • Completely routed • Circuit meets specified performance • Download configuration file (bit-steam) to the FPGA
Applications on FPGA: Low-level Implement Simulate Synthesize Map Place Route Download
Applications on FPGA: Low-level Implement Simulate Synthesize Map Place Route Download
Place • Bind each mapped resource to a physical device location • User Guided Layout (Chapter 16:Reconfigurable Computing) • General Purpose (Chapter 14:Reconfigurable Computing) • Simulated Annealing • Partition-based • Structured Guided (Chapter 15:Reconfigurable Computing) • Data Path based • Heuristics used • No efficient means for finding an optimal solution
Place (High-level) Netlist from technology mapping in A in B in C RAM E DFF F LUT D DFF G clk out
Place (High-level) Netlist from technology mapping FPGA physical layout I/O I/O I/O I/O in A in B in C I/O BRAM I/O LUT LUT RAM E I/O I/O DFF F LUT D LUT I/O I/O LUT BRAM I/O LUT I/O LUT DFF G I/O I/O clk LUT I/O LUT I/O out I/O I/O I/O I/O
Place (High-level) Netlist from technology mapping FPGA physical layout clk in C out I/O in A in B in C In A E I/O LUT G D F RAM E In B I/O DFF F LUT D LUT I/O I/O LUT BRAM I/O LUT I/O LUT DFF G I/O I/O clk LUT I/O LUT I/O out I/O I/O I/O I/O
Place (User-Guided) • User provide information about applications structure to help guide placement • Can help remove critical paths • Can greatly reduce amount of time for routing • Several methods to guide placement • Fixed region • Floating region • Exact location • Relative location
Place (User-Guided): Examples Part of Map Netlist DFF F LUT D DFF G FPGA Fixed region
Place (User-Guided): Examples Part of Map Netlist DFF F SDRAM LUT D DFF G FPGA Fixed region
Place (User-Guided): Examples Floating region FPGA Softcore Processor
Place (User-Guided): Examples Part of Map Netlist DFF F LUT D DFF G FPGA Exact Location LUT BRAM LUT LUT LUT LUT LUT LUT LUT BRAM LUT LUT LUT LUT LUT LUT LUT LUT
Place (User-Guided): Examples Part of Map Netlist DFF F LUT D DFF G FPGA Exact Location LUT BRAM LUT LUT G LUT D F LUT LUT LUT BRAM LUT LUT LUT LUT LUT LUT LUT LUT
Place (User-Guided): Examples Part of Map Netlist G DFF F D F LUT D DFF G FPGA Relative Location LUT BRAM LUT LUT LUT LUT LUT LUT LUT BRAM LUT LUT LUT LUT LUT LUT LUT LUT
Place (User-Guided): Examples Part of Map Netlist G DFF F D F LUT D DFF G FPGA Relative Location LUT BRAM LUT LUT LUT LUT LUT LUT LUT BRAM LUT LUT LUT LUT LUT LUT LUT LUT
Place (User-Guided): Examples Part of Map Netlist G DFF F D F LUT D DFF G FPGA Relative Location LUT BRAM LUT LUT LUT LUT LUT LUT LUT BRAM LUT LUT LUT LUT LUT LUT LUT LUT
Place (General Purpose) • Characteristics: • Places resources without any knowledge of high level structure • Guided primarily by local connections between resources • Drawback: Does not take explicit advantage of applications structure • Advantage: Typically can be used to place any arbitrary circuit
Place (General Purpose) • Preprocess Map Netlist using Clustering • Group netlist components that have local conductivity into a single logic block • Clustering helps to reduce the number of objects a placement algorithm has to explicitly place.
Place (General Purpose) • Placement using Simulated Annealing • Based on the physical process of annealing used to create metal alloys
Place (General Purpose) • Simulated annealing basic algorithm • Placement_cur = Inital_Placement; • T = Initial_Temperature; • While (not exit criteria 1) • While (not exit criteria 2) • Placement_new = Modify_placement(Placement_cur) • ∆ Cost = Cost(Placement_new) – Cost(Placement_cur) • r = random (0,1); • If r < e^(-∆Cost / T), Then Placement_cur = Placement_new • End loop • End loop
Place (General Purpose) • Simulated annealing: Illustration
Place (General Purpose) • Placement using partitioning
Place (Structured-based) • Automatically leverage structure of the application • Algorithms my work well for a give structure, but will likely give unacceptable results for an design with little regular structure.
Place (Structure-based) • Datapath-based example
Next Lecture • Finish Placing Applications on FPGAs • Compute Models: Part I • Recommended Reading • Design Patterns for Reconfigurable Computing • http://ic.ese.upenn.edu/abstracts/despat_fccm2004.html
Questions/Comments/Concerns • Write down • Main point of lecture • One thing that’s still not quite clear • If everything is clear, then give an example of how to apply something from lecture OR