Questions on tree-inline.c

[Jason Merrill]

On Fri, 13 Jun 03 09:00:47 EDT, kenner@vlsi1.ultra.nyu.edu (Richard Kenner) wrote:

>     This problem should go away when tree-ssa is merged into mainline. 
>
> Well that may be, but I'm still quite curious how we have the mess we have
> now and what to do about it in the mainline.  As you know, I view
> tree-ssa as an experiment, which may or may not turn out to be helpful.
> So I'd like to get this cleaned up in the meantime.

The way to clean it up is to have a language-independent functions-as-trees
representation.  Which only exists on the tree-ssa branch.  I suppose you
could try to merge some of the gimplifier code onto the trunk, but that
seems premature.

Jason