N. Togawa et al., AN FPGA LAYOUT RECONFIGURATION ALGORITHM-BASED ON GLOBAL ROUTES FOR ENGINEERING CHANGES IN SYSTEM-DESIGN SPECIFICATIONS, IEICE transactions on fundamentals of electronics, communications and computer science, E81A(5), 1998, pp. 873-884
Citations number
26
Categorie Soggetti
Engineering, Eletrical & Electronic","Computer Science Hardware & Architecture","Computer Science Information Systems
Rapid system prototyping is one of the main applications for field-pro
grammable gate arrays (FPGAs). At the stage of rapid system prototypin
g, design specifications can often be changed since they cannot be det
ermined completely. In this paper, layout design change is focused on
and a layout reconfiguration algorithm is proposed for FPGAs. The targ
et FPGA architecture is developed for transport processing. In order t
o implement more various circuits flexibly it has three-input lookup t
ables (LUTs) as minimum logic cells. Since its logic granularity is fi
ner than that of conventional FPGAs, it requires more routing resource
s to connect them and minimization of routing congestion is indispensa
ble. In layout reconfiguration, the main problem is to add LUTs to ini
tial layouts. Our algorithm consists of two steps: For given placement
and global routing of LUTs, in Step 1 an added LUT is placed with all
owing that the position of the added LUT may overlap that of a preplac
ed LUT; Then in Step 2 preplaced LUTs are moved to their, adjacent pos
itions so that the overlap of the LUT positions can be resolved. Globa
l routes are updated corresponding to reconfiguration of placement. Th
e algorithm keeps routing congestion small by evaluating global routes
directly both in Steps 1 and 2. Especially in Step 2, if the minimum
number of preplaced LUTs are moved to their adjacent positions, our al
gorithm minimizes routing congestion. Experimental results demonstrate
that, if the number of added LUTs is at most 20% of the number of ini
tial LUTs, our algorithm generates the reconfigured layouts whose rout
ing congestion is as small as that obtained by executing a conventiona
l placement and global routing algorithm. Run time of our algorithm is
within approximately one second.