hmmm. Lets see. We could definitely make something useful out of that chemical garbage precursor you kids call methamphetamine :P ;D .
If we attach a three carbon chain to the meth. amine group, we have a perfect precursor for an acyclic fentanyl analogue. That attached carbon chain must of course terminate with a leaving group, preferably bromine in this case. Now all we have to do is alkylate some aniline with that halogenated compound and acylate the aromatic amine with propionyl chloride. Aniline can be easily alkylated by employing a Pd salt/ligand catalyst along with a few equivalents of strong base (K-t-BuO).
Another synth would start with acylation of the methamphetamine amine with 3-chloropropionyl chloride then alkylate and acylate as mentioned above. The final step will be reducing the carbonyl group attached to the phenylisopropylamine nitrogen. The product N-(3 methyl-phenylisopropylamino)-propyl-N'phenyl propionamide appears similar to some structures on this page:
http://www.ch.ic.ac.uk/ectoc/echet98/pub/028/index.htm (http://www.ch.ic.ac.uk/ectoc/echet98/pub/028/index.htm)
So what do you know, turns out that meth isn't completely useless afterall! :P