Re: A newbie's view

["Seth M. Landsman" <seth@job.cs.brandeis.edu>]

On Fri, Jan 29, 1999 at 02:53:26AM +0100, Kenneth Skaar wrote:
> 
> 
> On Thu, 28 Jan 1999, Seth M. Landsman wrote:
> 
> > 	This definitely shouldn't be part of the kernel proper (i.e., the
> > kernel that is compiled should be identical after these changes go in).
> > But I do think things like this should be part of the kernel makefile ...
> > 	I think for any complex system, it becomes increasingly necessary
> > to have some sort of good configuration and testing facility.  I think
> > that the kernel is lacking in both.
> >
> All of this is currently beeing worked on by the Linux Kernel Manager
> project(http://www.caribdata.co.uk/psi-domain/lkm.html). They haven't been
> around for long, but they are moving forward, and it seems like their
> doing The Right Way(tm). I dont know how they plan to keep up with all the
> new configuration-options that pops up now and then, but I guess they'll
> read and parse the source config-files, and then offer wizard-style help
> for the "known" parts, and just plain kernel-build-help for the new stuff. 

	It does not seem like this is being worked on by the LKM project,
after looking through the web page.  

	Okay, here's a list of what I want :

o grok from a currently running kernel the entire .config or pieces of it
(this has been talked to death on linux-kernel)

o grok from hardware probing various important addresses (i.e., I/O of the
sound card)

o a more reasonable set of defaults (perhaps based on above hardware
probing)

o big obvious warnings if the user turns off a "necessary to survive"
option, like ELF support

o big obvious warnings if the user turns off a driver that was detected in
hardware or turns on a driver that was not.

o some sort of `make test` that make sure the kernel is valid and has a
decent chance of booting the machine in question.

	It doesn't seem that LKM does most of these.  

	Furthermore, anything that is done *SHOULD WORK FROM THE
COMMANDLINE*.  I want to ssh to my router and recompile a kernel without
having to forward an X session.  LKM might be something we can merge with
or can use whatever is done, but LKM itself is unacceptable because it is
only X based.

> Anyway, IMHO keeping a "user" app like this out of the kernel source
> tree is a _good_ idea. 

	Some of this would be such an integral part of the build process
that it would have to be in the source tree, I think.

-Seth
--
"It is by will alone I set my mind in motion"
-
Linux-future: thinking about the future of the Linux kernel
http://humbolt.nl.linux.org/lists/