[klibc] curious unpacking issue

[Aaron Griffin]

On 4/10/06, H. Peter Anvin <hpa at zytor.com> wrote:
> Aaron Griffin wrote:
> > Yeah, I did that for a while - however, I'd like to get this sort of
> > stuff done as soon as possible, so I was trying to get it compatible
> > with the current tree.  Also, the whole git pull is huge compared to
> > the kernel.org tarballs - though it's not huge issue, I'm running a
> > tad low on diskspace, heh.
>
> Yes, but you also are trying a combination of things that has never been
> tested, and thus, you're completely on your own.  I'm not sure what you
> mean with "getting this sort of things done as soon as possible"; I'm
> currently iterating pretty tightly with Andrew Morton trying to get
> klibc into -mm.

Well, I kind of misspoke.  Let me rephrase a tad - I'm planning on
replacing an initrd boot scheme with an initramfs one, but not
1-for-1, rather it would be 'nicely' designed, taking advantage of the
overwriting scheme.
I guess the "soon as possible" mention really has to do with getting
it working in the vanilla kernel, as I am unsure on the timeline for
inclusion.  Once klibc is moved to -mm, it should be quicker to move
to the vanilla kernel.

For now I will attempt with the git tree.  Assuming this works, I may
end up diff-ing the init/ code for the vanilla tree, just to see where
it stands.

Yeah, I know I'm trying things that haven't been tested, but, in a
way, I enjoy that.

- Aaron