6.16 gcc omit-frame pointer

Stuart Stegall keltor at keltor.org
Wed Dec 1 13:23:54 PST 2010


On Wed, Dec 1, 2010 at 2:42 PM, Gilles Espinasse <g.esp at free.fr> wrote:
>
> ----- Original Message -----
> From: "David Jensen" <djensenlin at windstream.net>
> To: <lfs-dev at linuxfromscratch.org>
> Sent: Wednesday, December 01, 2010 9:01 PM
> Subject: Re: 6.16 gcc omit-frame pointer
>
>
>> On Wed, 01 Dec 2010 11:38:06 -0600
>> Bruce Dubbs <bruce.dubbs at gmail.com> wrote:
>>
>> > David Jensen wrote:
>> >
>> > > The point of the thread is that the 'chapter 5 gcc pass 2' and the
>> > > 'chapter 6 gcc' instructions and explanations are inconsistent,
>> > > confusing and wrong for all but i?86.  Maybe even completely
>> > > outdated.
>> >
> Right for the difference for non i?86
> The chap5 gcc pass2 text say "For x86 machines, " but the instruction miss
> the
> case `uname -m` in
>  i?86)esac
> that is present in chap6
>
> The difference from i?86 and x86_64 is normal, that 's explained in
> http://www.mail-archive.com/lfs-dev@linuxfromscratch.org/msg02080.html
>
> I don't know if gcc-4.5 still behave that way.
>
> Gilles
>
> --
> http://linuxfromscratch.org/mailman/listinfo/lfs-dev
> FAQ: http://www.linuxfromscratch.org/faq/
> Unsubscribe: See the above information page
>

It's a fix that's only applicable to x86.  amd64/arm is always
-fno-frame-pointer (unless of course you are explicit with turning it
on.)  gcc-4.6 will change x86's behavior to -fno-frame-pointer by
default.  Honestly, I thought this issue was fixed in gcc-4.3.



More information about the lfs-dev mailing list