Re: [PATCH] Repair TYPE_REF_CAN_ALIAS_ALL machinery

["Richard Guenther" <richard dot guenther at gmail dot com>]

On 5/22/06, Eric Botcazou <ebotcazou@adacore.com> wrote:
> > I wonder why the machinery for handling pointers with alias set zero
> > does not trigger
> > here.  Can you check if the ada langhook for get_alias_set returns
> > something different
> > than -1 for pointers with TYPE_REF_CAN_ALIAS_ALL?  Or why get_alias_set
> > does not return 0 for those types?
>
> I'm under the impression that you're confusing alias set of the pointed-to
> type and alias set of the pointer type.  As suggested by Olivier, another
> possible track could be to tweak tree-ssa-alias.c:may_alias_p, although I'm
> not sure whether this would be sufficient.

I don't like the tree-ssa-operands.c part of the patch as we should
have a mechanism
that deals with this problem already, as we have char* pointers.  Btw.
I couldn't construct
a C testcase that failed.  So, as you noticed I confused the type
somewhat so one would
need to amend the get_alias_set calls in tree-ssa-alias.c:get_tmt_for
and friends to
check for TYPE_REF_CAN_ALIAS_ALL on the pointer type and use alias set 0 in this
case.

Richard.