[ tree-ssa] PRE problem

[law@redhat.com]

In message <1077307394.3169.188.camel@localhost.localdomain>, Diego Novillo wri
tes:
 >On Fri, 2004-02-20 at 14:56, law@redhat.com wrote:
 >
 >>   p = x;
 >>   y = p;
 >>       *y;
 >> 
 >> "y" will have a tag, but by no means does that mean p or x will have a
 >> tag.
 >> 
 >We are talking past each other again.
It wouldn't be the first time.

 > When I say it's impossible for
 >them to have different tags, I am talking about *non-NULL* tags.  It's
 >perfectly OK for a pointer to *not* have a tag.  Only dereferenced
 >pointers need tags.
Now this statement I would believe!  And it's fundamental to the problem
at hand.   So just to be sure, we agree that we can have a copy between
pointers, one of which has a tag and one of which does not?


jeff