It's better to focus on low-level optimizations within LLVM IR I think,
high-level optimizations would better be done beforehand. It's not a bad
marriage though. :)

-- 
Eray Ozkural, PhD candidate.  Comp. Sci. Dept., Bilkent University, Ankara
http://groups.yahoo.com/group/ai-philosophy
http://myspace.com/arizanesil http://myspace.com/malfunct