Subject: Re: RFC: migration to a fully dynamically linked system
To: Todd Vierling <firstname.lastname@example.org>
From: Bill Studenmund <email@example.com>
Date: 12/29/2001 15:07:26
On Sat, 29 Dec 2001, Todd Vierling wrote:
> Perhaps the question should be rephrased as, "do we *need* the ability for
> init to have external auth methods, and so forth?" If not, then init can
> stay static without pain.
I would vote no.
Mainly as I think init is special. After we have init running, we have the
standard UNIX proces model in full effect. Before init is up, we don't. If
init can't come up, we can't use the restore tools. :-( So let's make sure
init will make it up.
Or actually, is there a way we can have a tool which you hand a dynamic
program (or blob) and it spits out a static one? If so, then I would vote
we ship init itself static, but ship a dynamic init, and a how-to saying
how to make a static one. The idea is to make a static init which pulls in
whatever modules we would want in a dynamic libc.
I realize that means changing the table of built-in stuff for libc. And it
could be a whole thread in and of its own. :-)