From: jle@ural.owlnet.rice.edu   
      
   =?ISO-8859-1?Q?Pertti_Kellom=E4ki?=   wrote:   
   >I am trying to wrap my head around the Banerjee inequality (a basis   
   >for a particular form of dependence testing in loops).  While I   
   >understand the gross outline, I am trying to work out the details to   
   >convince myself. However, the proofs in Allen and Kennedy's Optimizing   
   >Compilers for Modern Architectures are given in such high level that I   
   >am having a hard time filling in some of the gaps.   
   >   
   >Does anyone know of sources where the proofs would be spelled   
   >out in more detail?   
      
     See:   
      Zima and Chapman, "Supercompilers for Parallel and Vector Computers", 1991.   
      Chapter 4 and Appendix B.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)