Re: [PATCH] Make alias_sets_conflict_p less conservative

["Daniel Berlin" <dberlin@dberlin.org>]

On Tue, Mar 4, 2008 at 6:55 PM, Richard Kenner
<kenner@vlsi1.ultra.nyu.edu> wrote:
> > This makes alias_sets_conflict_p less conservative.  rev 34373 (Kenner)
>  > added a check that makes a structure containing a member with alias
>  > set zero alias everything.  Which IMHO is non-obvious and pessimizes
>  > for example the testcase in PR27799.  Richard, can you explain this
>  > change?
>
>  I'm not certain, but I think it goes like this: if a structure has
>  a member with alias set X (nonzero), we normally mark the alias sets
>  in such a way that the alias set of the structure as a whole conflicts
>  with alias set X.  But if X is zero, the subsetting mechanism doesn't
>  allow doing that.  So we have to do it explicitly, like in this code.
>
>  In other words, this is exactly analogous to the non-zero alias set
>  of a field case.
>

Sadly, you are right about how the subsetting mechanism works, but he
is right in what the result we want should be.

This is because the the type based rules (at least for C) say "access
through pointer char can validly alias anything" but not "access
through pointer to anything can validly alias char".

Because alias_sets_conflicts_p checks both ways, it ends up claiming
the second statement is true as well.

In reality, you need a directional aliasing test here, not
alias_sets_conflict_p (IE you want something that says "can a pointer
of this point to this type over there)

--Dan