This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Make alias_sets_conflict_p less conservative
On Wed, 5 Mar 2008, Richard Kenner wrote:
> > Err, right. But tree aliasing is computing a representation for TBAA,
> > using alias set conflict queries but _not_ looking at individual accesses.
>
> That seems quite wrong to me.
>
> > So, is alias_set_subset_of () wrong as currently implemented? Does
> > it lack the ase->has_zero_child test in your opinion? Otherwise I'll
> > try using that for these queries.
>
> I don't know enough to answer those questions.
That is sad, because you knew enough to fix exactly the same code
in alias_sets_conflict_p.
> What I *do* know is that if you have x.y, with y addressable, any code
> that looks at the alias set of X *for any reason* is broken.
Well, you obviously do not have enough knowledge of the alias
representation used in tree-ssa and I won't try to start to explain.
If you want, take the testcase in PR27799, build with
-O2 -fdump-tree-salias-vops-details and look at the 052t.salias dump.
Richard.