This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [2/7] v7: Handle frame pointer for stack alignment
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: "Ian Lance Taylor" <iant at google dot com>
- Cc: gcc-patches at gcc dot gnu dot org, ian at airs dot com, rguenther at suse dot de
- Date: Mon, 30 Jun 2008 10:02:26 -0700
- Subject: Re: [2/7] v7: Handle frame pointer for stack alignment
- References: <20080628180825.GC21773@lucon.org> <m363rqx5yi.fsf@google.com>
On Mon, Jun 30, 2008 at 9:06 AM, Ian Lance Taylor <iant@google.com> wrote:
> "H.J. Lu" <hjl.tools@gmail.com> writes:
>
>> Frame pointer register is used by stack alignment to save and restore
>> stack pointer. So we have to eliminate frame pointer when stack is
>> aligned. We also can only set if frame pointer is needed in one place.
>> This patch moved frame_pointer_needed initialization to compute_regsets
>> which is called before init_elim_table. Tested on Linux/ia32,
>> Linux/x86-64 and Linux/ia64. Ian, Ricard, is this OK for trunk?
>
> There have been so many different versions of these patches that I
> can't keep track of them. I know that you didn't test this patch in
> isolation, because it refers to fields which aren't in trunk.
>
> It's basically OK with me to move the initialization of
> frame_pointer_needed, but I would like to see a patch for that which
> does not mix in the stack alignment stuff. Then I would ideally like
> to see a complete stack alignment patch, not a patch which mixes stack
> alignment changes with other changes.
>
Hi Ian,
The separate frame pointer initialization patch is at
http://gcc.gnu.org/ml/gcc-patches/2008-06/msg01269.html
I have been testing it on Linux/ia32, Linux/x86-64 and Linux/ia64
without any stack alignment patches. OK for trunk?
Thanks.
--
H.J.