On Tue, 12 Dec 2000, Luke Mewburn wrote:

# Is there any reason not to provide this functionality until a separate
# kernel configuration mechanism exists? (which hasn't even been
# discussed let alone had an implementation started or completed :)

I didn't see anything about a separate configuration mechanism.  I
consider 'config' to be quite adequate, actually, though it might be
nice to be able to modularise one's setup and say config file file file
file master-file without having to populate the master-file with .includes.

Regarding putting the config in the kernel would be cool, accessible via
a (possibly restricted) sysctl (though I can't see what security would be
broken via a kernel config file).

Possibly available as /kern/config?  (along the same lines as /kern/syms
or whatever for kernel symbols (which REALLY _should_ be posted out by the
bootloader to said destination for easy retrieval by ps/libkmem/gdb-post-
mortem/dmesg, but that's another story)).

# 	``oh no, don't do <foo>, we'll have feature <bar> to do that better''

Bah.  "Let's do <foo> now.  if <bar> materialises, AND it *really* works
better, then do <bar>.  If not, we sill have <foo>."

Which, I think, you said in the paragraphs elided below.

# Luke.

				--*greywolf;
--
*BSD: Microsoft ask you where you want to go.  BSD gets you there.