This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [tree-ssa] AST optimizer in C++?


On Mon, 26 Aug 2002, Daniel Berlin wrote:

> On Mon, 26 Aug 2002, Chris Lattner wrote:
> 
> > On Sun, 25 Aug 2002, Daniel Berlin wrote:
> > 
> > > > Is there a reason that you'd like to have a SIMPLE form that is not in
> > > > SSA?
> > >
> > > Um, yes.
> > > It artificially limits us to that representation.
> > > Say later, someone wants to use some other form of SSA, or non-SSA
> > > (dependence flow graph, etc).
> > > If you make it actually part of the trees, they now have to care about
> > > it.
> > > If it's just annotations, only what wants to look at those annotations
> > > look at it.
> > 
> > Actually, that's not true at all.  Using SSA as the base representation is
> > just like using a conventional register based underlying representation:
> > you can still layer other representations ON TOP of SSA, you just need to
> > keep the SSA up-to-date as you transform the layered representation (just
> > as you must do with the register based representation).
> 
> No kidding, which is the problem.

Just to be clear, i mean the fact that we have to layer things on top of 
it is the problem. not the updating.



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]