This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [C++ PATCH] Fix aliasing warnings for typeid (PR c++/32260)


On 11/1/07, Jakub Jelinek <jakub@redhat.com> wrote:
> On Thu, Nov 01, 2007 at 01:53:46PM +0100, Richard Guenther wrote:
> > The comment hints at that this should be
> >
> >        gcc_assert (! TYPE_ALIAS_SET_KNOWN_P (pseudo_type_info));
> >
> > or does this break at any time?
>
> Sure, typeid_ok_p is called many times (once for each typeid parsed
> at least) and once we set it it will be TYPE_ALIAS_SET_KNOWN_P
> (pseudo_type_info).

Ah, I see.

> I could perhaps add some static bool variable, so that this would be done
> only once.  But is it worth it?

Or

+  if (! TYPE_ALIAS_SET_KNOWN_P (pseudo_type_info))
+    {
+      tree type_info_type = TYPE_MAIN_VARIANT (const_type_info_type_node);
+      TYPE_ALIAS_SET (pseudo_type_info) = get_alias_set (type_info_type);
+    }
   else
      gcc_assert (TYPE_ALIAS_SET (pseudo_type_info) == get_alias_set
(type_info_type));

I think it's worth it for checking purposes.

Richard.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]