> You might be interested in the Ph.D. thesis of Andrzej Bednarski, and   
   > in his other publications, and in papers he cites.   
   >   
   > - anton   
      
   Yes, that's probably a good start. Their work (materialized as the   
   OPTIMIST compiler prototype) attacks all three problems (code   
   selection, register allocation and instruction scheduling) by means of   
   dynamic and integer linear programming.   
   You understand of course, that the specific solution might be too slow   
   (serious number in orders of magnitude) for your purposes.   
      
   Nikolaos Kavvadias   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)