Re: [RFC] klibc requirements, round 2

[Tom Rini]

On Thu, Jan 10, 2002 at 05:50:52PM -0800, Torrey Hoffman wrote:
> Tom Rini wrote:
> > On Thu, Jan 10, 2002 at 03:18:49PM -0800, Greg KH wrote:
> ... 
> > > 	- image viewer
> > > 	- mkreiserfs
> > 
> > I think these are examples of misunderstanding what initramfs _can do_
> > with what we (might) need a klibc to do.  
> ...
> > These programs _might_ compile with a klibc, but I wouldn't 
> > worry about
> > it.  uClibc is what should be used for many of these custom 
> > applications
> 
> Well, as the person who first mentioned mkreiserfs and the like,
> I agree with you.  For the majority of systems out there which 
> aren't using initrd now, a minimal klibc for an unmodified 
> initramfs makes sense.

Okay.

> My concern is with the minority who are using initrd, and in
> some cases a very customized initrd.  

Right.  And moving that very customized initrd over to an initramfs
should be painless, once the kinks/bugs are worked out of the yet-to-be
created programs that exist in the kernel now.

> The important thing, I think, is that it should be easy for
> less-than-guru level hackers to add programs to the initramfs,
> even if the program they want can't be linked with klibc.
> 
> This really comes down to: What will the build process be for
> these initramfs images?

It's a cpio archive, occording to the draft spec hpa posted.

> By the way, is initramfs intended to supercede initrd, or will 
> they co-exist?  

I _think_ co-exist.

-- 
Tom Rini (TR1265)
http://gate.crashing.org/~trini/