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, Mar 5, 2008 at 11:55 AM, Richard Guenther <rguenther@suse.de> wrote:
>
> On Wed, 5 Mar 2008, Daniel Berlin wrote:
>
> > On Wed, Mar 5, 2008 at 11:21 AM, Richard Guenther <rguenther@suse.de> wrote:
> > >
> > > Which means like the following (untested) patch, also fixes PR27799.
> > >
> > > Richard.
> > >
> >
> > This looks right to me because we always know which one is the pointer
> > and which one is the pointed-to var.
> > So it's fine assuming it passes bootstrap and regtest
>
> It doesn't work.
>
You reversed the arguments.
alias_set_subset_of tells you if the first is a subset of the second.
So you want to know if the var is a subset (actually subset or equal
to) of the ptr.
The way the arguments are in your patch, you are asking if ptr is a
subset of var.
>