This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Incorrect bitfield aliasing with Tree SSA
- From: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- To: dberlin at dberlin dot org
- Cc: anemet at caviumnetworks dot com, ebotcazou at libertysurf dot fr, gcc-patches at gcc dot gnu dot org
- Date: Sun, 17 Jun 2007 07:50:43 EDT
- Subject: Re: Incorrect bitfield aliasing with Tree SSA
- References: <18034.48960.772642.116034@localhost.localdomain> <4aca3dc20706151412m6bac4d74lec2cc3e363503184@mail.gmail.com> <18035.3033.118272.195606@localhost.localdomain> <4aca3dc20706151623t2e85b196ga805e6c30519924c@mail.gmail.com> <18036.42952.55717.393712@localhost.localdomain> <4aca3dc20706162035w1efd334cl86aeaa26da7f5fc9@mail.gmail.com>
> This is fine.
>
> Any "value" we can get from DECL_NONADRESSABLE_P, we will get through
> handling it explicitly when pruning.
>
> > Adam
> >
> > * alias.c (record_component_aliases): Call record_alias_subset for
> > DECL_NONADDRESSABLE_P fields too.
No. Once you've subsetted it in this way, you lose most of the value
of it! The main reason for the flag is precisely for the usage that is
proposed being eliminated here.