Perception of LFS

Greg Schafer gschafer at
Sun Jan 12 04:01:25 PST 2003

On Sun, Jan 12, 2003 at 12:40:38PM +0100, Matthias Benkmann wrote:
> > Alan Cox says:
> > 
> > "I get so many weird never duplicated reports from linux from scratch
> > people that don't happen to anyone else that I treat them with deep
> > suspicion. Especially because it sometimes goes away if they instead
> > build the same kernel with Debian/Red Hat/.. binutils/gcc"

<snip stuff I agree with>

> I don't think this is at all necessary. I think that most problems are
> caused by the mistakes people make when building LFS. We can't do anything
> about those. Even if we instructed people to build the kernel with
> gcc-2.95, many people would use gcc-3.x to do it and YOU BET that many of
> those wouldn't bother to include this information in a support request,
> often on purpose to avoid being told to recompile (which would take
> *their* time instead of the time of the time of the people on lfs-support,
> and that absolutely must no be).

Yes, I agree. But if we want to be taken seriously by upstream developers
then we really should be doing things "by the book" and I'm not talking
about the LFS book :-) Hence why I'm always on about improving technical
standards and the way we do things. That way, if the LFS book gets it right,
then the silly support requests will diminish.

> BTW, you (Greg) are as guilty in all these respects as someone can be.
> Think about how your sed couldn't handle binaries and it was only a couple
> posts later that you came up with the info that it was not built according
> to the book.

Bad example. One of the so called distros that AC trusts (Debian) built
their sed exactly the same way.

> Think about the loadkeys issue (which some people may well
> have reported as a linux kernel bug), wasn't this caused by one of the
> hacks/tweaks on your page?

Another bad example. I put it to you that the use of my tweak exposed a bug
in gcc-3.2 compiling at -march=i686 which eventually led to it getting fixed
by the gcc developers and thus saved pain for millions of Linux users all
across the globe, including you!

> Please, don't take this the wrong way. LFS is made for tweaking. It's the
> spirit of LFS, and you are doing a very valuable service to the LFS
> community. However, because of what you do, you are the LAST person on
> earth to have a right to criticize Alan Cox for refusing to deal with
> support requests from LFS users.

You missed my point. AC is the last person on earth I would criticize. The
point is that Linux God's like him do not trust LFS and that is a real
problem IMHO.

Unsubscribe: send email to listar at
and put 'unsubscribe lfs-dev' in the subject header of the message

More information about the lfs-dev mailing list