I do have used this program but remember creating input files was difficult (I wrote an Excel macro ultimately). There is an alternative program using Bayesian approaches which is more handy (BAYESCAN, Foll M, Gaggiotti O: A genome-scan method to identify selected loci appropriate for both dominant and codominant Markers: A Bayesian perspective. Genetics 2008, 180(2):977-993.).
But if you are still interested in the DFDist I may search in my files where I have an example for the input.
The program FAMD can make the dfdist input file for you from a typical spreadsheet with loci in columns and individuals in rows (or vice versa). http://www.famd.me.uk/famd.html
It is also fairly easy to manipulate data sets so that you can do dfdist comparisons among individual locations (e.g., Egan et al. 2008, Evolution 62:1162-1181).