This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug middle-end/35705] [4.3/4.4 Regression] Symbol address check eliminated by C frontend.
- From: "rguenther at suse dot de" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 27 Mar 2008 09:52:53 -0000
- Subject: [Bug middle-end/35705] [4.3/4.4 Regression] Symbol address check eliminated by C frontend.
- References: <bug-35705-11932@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #12 from rguenther at suse dot de 2008-03-27 09:52 -------
Subject: Re: [4.3/4.4 Regression] Symbol address check
eliminated by C frontend.
On Thu, 27 Mar 2008, jakub at gcc dot gnu dot org wrote:
> ------- Comment #11 from jakub at gcc dot gnu dot org 2008-03-27 09:42 -------
> That will unnecessarily punish the majority of targets which use simple
> pointers.
> I thought you can align functions more than FUNCTION_BOUNDARY by
> ASM_OUTPUT_MAX_SKIP_ALIGN, but apparently that's not unconditional.
> In that case, we should have a target macro or target hook, which will tell us
> alignment of ADDR_EXPR <FUNCTION_DECL>, defaulting to DECL_ALIGN_UNIT of the
> FUNCTION_DECL, and get_pointer_modulus_and_residue should use that.
I don't think alignment checks on function pointers are common
(nor portable, a function "pointer" is an opaque object). So a new
target hook is overkill IMHO.
Richard.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35705