Factoring algorithms on rtl ?

[Gábor Lóki]

Joern Rennecke wrote:
>>We would very much like to continue this work, because even in this very
>>simple code motion algorithm we have extra gain beside the mentioned
>>algorithm when optimizing for size. I expect significant reductions from
>>the more sophisticated factorings. Do you support this?
> 
> 
> I wonder how to make sure that local registers are actually local to
> the common sequences.  Do we have accurate REG_DEAD notes at the
> point you want to do this factoring?

No. The patch uses the def-use info to decide if the change could be
done.