On Thu, Sep 24, 2020 at 1:26 PM Jan Hubicka wrote:
>
> > >
> > > I will do (but need to think bit of a redundancy between comment in
> > > ipa-modref and ipa-modref-tree)
> >
> > One place is enough - just add a pointer to the other place.
> Here is updated patch I am testing. I adds documentatio
> >
> > I will do (but need to think bit of a redundancy between comment in
> > ipa-modref and ipa-modref-tree)
>
> One place is enough - just add a pointer to the other place.
Here is updated patch I am testing. I adds documentation into
ipa-modref-tree.h that is perhaps more natural place and l
On Thu, Sep 24, 2020 at 12:54 PM Jan Hubicka wrote:
>
> > > + else if (TREE_CODE (op) == SSA_NAME
> > > + && POINTER_TYPE_P (TREE_TYPE (op)))
> > > +{
> > > + if (DECL_P (base) && !ptr_deref_may_alias_decl_p (op, base))
> > > + return false;
> > > + if (TREE_CODE (bas
> > + else if (TREE_CODE (op) == SSA_NAME
> > + && POINTER_TYPE_P (TREE_TYPE (op)))
> > +{
> > + if (DECL_P (base) && !ptr_deref_may_alias_decl_p (op, base))
> > + return false;
> > + if (TREE_CODE (base) == SSA_NAME
> > + && !ptr_derefs_may_alias_p (op, base))
On Thu, Sep 24, 2020 at 11:06 AM Jan Hubicka wrote:
>
> Hi,
> this patch re-adds tracking of accesses which was unfinished in David's patch.
> At the moment I only implemented tracking of the fact that access is based on
> derefernece of the parameter (so we track THIS pointers).
> Patch does not
