[PATCH][RFC] A GIMPLE and "GIMPLE type system" verifier

[Richard Guenther]

On 6/1/07, Richard Kenner <kenner@vlsi1.ultra.nyu.edu> wrote:
> > Though, it's not clear to me how you could get away with pointer
> > equality of types (IE share types), and still be able to use TBAA info
> > for each type (without seriously affecting the precision of TBAA).
>
> That's true.  Pointer to A and pointer to type-equivalent-to-A can't be
> treated as the same.

You'd need to have the frontends tag each memory reference it produces
with a TBAA tag and provide means to test if two such tags conflict.

Richard.