Will tree-ssa be GCC 3.5?

[Robert Dewar]

Richard Kenner wrote:
>     Right, quite the opposite really. These would normally be regarded as 
>     very high-level optimizations, and I do think it is right that Fortran 
>     has special requirements here.
> 
> By "high-level", I meant the sort of interprocedural stuff that has been
> talked about for C++.

Yes, but the class of optimizations being talked for Fortran are of a 
quite different nature.