On 6/10/2005 12:16:07, Richard Freeman ([EMAIL PROTECTED]) wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Daniel Gryniewicz wrote:
> > 
> > The big reason would be because gcc 3.3.x (the stable compiler on x86)
> > doesn't support it.  It has a patch that adds the option to gcc, but it
> > does nothing.  Until x86 is on 3.4.x by default, you can't expect full
> > support for stack-protector.
> > 
> 
> I hadn't even realized that.  And I thought that amd64 tended to lag
> behind!  As I recall 3.4 had quite a few benefits, as I normally don't
> tend to run bleeding-edge...

However Daniel's statement is false.  ssp is included for x86 in gcc-3.3.x, and 
has been for a long time.  You can't see it so easily on most 3.3.x versions as 
they're missing the definition of macros __SSP__ and __SSP_ALL__ but it's there 
nonetheless.

SSP is currently not in any of the 4.0 versions; work is still ongoing there 
(SSP for 4.0 is back-ported from 4.1, where it will be included upstream).

$ (cd /usr/portage/sys-devel/gcc; grep SSP_STABLE *ebuild)
gcc-3.3.5-r1.ebuild:SSP_STABLE="x86 sparc amd64"
gcc-3.3.5.20050130-r1.ebuild:SSP_STABLE="x86 sparc amd64"
gcc-3.3.5.20050130-r2.ebuild:SSP_STABLE="x86 sparc amd64"
gcc-3.3.6.ebuild:SSP_STABLE="x86 sparc amd64"
gcc-3.4.3-r1.ebuild:SSP_STABLE="x86 sparc amd64 ppc ppc64"
gcc-3.4.3.20050110-r2.ebuild:SSP_STABLE="x86 sparc amd64 ppc ppc64"
gcc-3.4.4-r1.ebuild:SSP_STABLE="x86 sparc amd64 ppc ppc64"

Kev.



-- 
[email protected] mailing list

Reply via email to