winter.txt

Archaic archaic at indy.rr.com
Tue Feb 3 14:40:29 PST 2004


On Tue, Feb 03, 2004 at 03:11:58PM -0500, Robert Connolly wrote:
> 
> Follow the cvs book with this hint. Unfortunately -fpie is slightly
> broken in the glibc-2.3.3 that is in the LFS cvs book. I tested this
> against glibc-2.3-20040129.

FYI, it looks like that will be too new of a cvs. Ryan's pinpointing the
cvs date and so far he's favoring around the 15th of January before the
2.3.4 symbols start being added.

> You also need a binutils that understands -pie, either HJL, or
> FSF-cvs. If you use FSF-cvs binutils you will have to comment out the
> sanity code in glibc/configure. grep -n -e "too old" configure, and
> comment out the whole test. HJL should work without that hassle. I
> don't have a relro patch for FSF binutils yet either, so HJL is
> reccommended.

Have you seen any signs of a release date for FSF? or fixes for the
sanity code being implemented? As far as HLFS and LFS goes, I'm really
starting to favor HJL.

With that said, can anyone provide a valid reason why HJL should *not*
be used here or in LFS?

> Only need one on these.
> http://cvs.mandrakesoft.com/cgi-bin/cvsweb.cgi/~checkout~/ \
>         SPECS/gcc/gcc33-pie.patch
> http://www.linuxfromscratch.org/patches/downloads/gcc/ \
>         gcc-3.3-pie-1.patch
> ftp://twocents.mooo.com/pub/hcc/pie/gcc-3.3-pie-2.patch

Huh? I can understand if the mandrake patch is the same as the one in
the patches section (minus the LFS-type header), but what's the diff
between the pie-1 and pie-2 patches you've created?

What's your take on grsec vs selinux With grsec's other functionality
being used, are you also leaning towards it's access controls or is
selinux more favorable?

-- 
Archaic

Morality is always the product of terror; its chains and
strait-waistcoats are fashioned by those who dare not trust others,
because they dare not trust themselves, to walk in liberty.

- Aldous Huxley




More information about the hlfs-dev mailing list