This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix PR41144
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Richard Guenther <rguenther at suse dot de>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Mon, 21 Dec 2009 12:56:17 -0800
- Subject: Re: [PATCH] Fix PR41144
- References: <alpine.LNX.2.00.0909061844560.28140@zhemvz.fhfr.qr>
On Sun, Sep 6, 2009 at 8:46 AM, Richard Guenther <rguenther@suse.de> wrote:
>
> This fixes PR41144 - the C++ FE is somewhat unlucky with the
> order it completes types during template substitution and
> messes up canonical types this way. ?The following patch ensures
> we're not recording types with structurall equality requirements
> in the canonical hash table.
>
> Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
>
> Richard.
>
> 2009-09-06 ?Richard Guenther ?<rguenther@suse.de>
>
> ? ? ? ?PR middle-end/41144
> ? ? ? ?* tree.c (build_array_type): Do not record types marked
> ? ? ? ?with structural equality in the canonical type hashtable.
>
> ? ? ? ?* g++.dg/torture/pr41144.C: New testcase.
This caused:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42447
H.J.