As promised: LFS compilation summary
Philipp Christian Loewner
bOrgph at web.de
Thu May 8 16:20:57 PDT 2008
On Thu, 08 May 2008 22:47:42 +0200, Ag. D. Hatzimanikas
<a.hatzim at gmail.com> wrote:
> Years ago, I suggested a whole chapter or a new book, for the Kernel
> related stuff [boot process (init ...), udev, filesystems, virtual
> machines, etc...].
> We did also found the man (Alexander) who could actually write and
> support this chapter/book but the community or a part of the community
> didn't want that change at the time, so it was rejected.
> I never stopped to say that it was a big fault.
Perhaps I just didn't say exactly what I meant:
I don't want loads of information about the kernel and it's features
(although it would be a great thing).
But as a user relatively new to customizable Linux (and I think
many new user will share my opinion) I was really surprised to see
such boot failures the first time. Just thought of something
like "Don't forget to enable all the file systems that you need
in the kernel." And/or "It may be that the system doesn't boot
properly and halts with an error message similar to 'Populating
/dev with device nodes... [Insert error message saying it was unable
to do so here]'. This is not a hint to a not properly functioning
udev installation, but points at missing kernel features. Do not
delete the kernel source directory after copying the bzImage in
the boot folder in order to be able to start the host system and
compile the kernel again with more features enabled."
Perhaps with a hint that we can't support a complete list of
configurations and needed features.
> Now I am in a position that I don't know if we should continue with LFS,
> as development doesn't simple happen anymore.
> The new gcc is out for months, binutils had a new version and we still
> don't know if we'll have a release and who is going to take this out
> and ... we still struggle in BLFS for a release for almost a year.
I think I know what you want to tell me. But on the other hand:
I never learned that much in just a few days than by reading the book
and compiling a system. This book is truly great work and very
good of teaching at least the basics of a linux system. It would
be a pity if development didn't continue for whatever reason.
Also, what exactly do you mean by "development doesn't simply happen
anymore"? Was it very different on this mailing list a few years ago?
More information about the lfs-dev