[LLVMdev] Regalloc Refactoring

David Greene greened at obbligato.org
Tue Apr 17 13:52:18 PDT 2007


David Greene wrote:
> Evan Cheng wrote:
> 
>> Obviously, smart heuristics can make a big difference here (estimated  
>> register pressures, etc.) But the more important thing is how the  
>> passes down stream can recover from the earlier mistakes. By this, we  
>> mean live range splitting and re-materialization. 
> 
> Can you explain this a little more?  Do you mean that neither live

Oops, copied the wrong part of the paragraph.  I meant to respond to this:

 > Unfortunately, neither of these are feasible given the current
 > infrastructure.

                                -Dave



More information about the llvm-dev mailing list