From owner-freebsd-hackers Thu Jul 11 12:48:57 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA29033 for hackers-outgoing; Thu, 11 Jul 1996 12:48:57 -0700 (PDT) Received: from mail.barrnet.net (mail.barrnet.net [131.119.246.7]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id MAA29028 for ; Thu, 11 Jul 1996 12:48:55 -0700 (PDT) Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by mail.barrnet.net (8.7.5/MAIL-RELAY-LEN) with SMTP id LAA15368 for ; Thu, 11 Jul 1996 11:29:07 -0700 (PDT) Received: (from narvi@localhost) by haldjas.folklore.ee (8.6.12/8.6.12) id VAA16323; Thu, 11 Jul 1996 21:30:42 +0300 Date: Thu, 11 Jul 1996 21:30:41 +0300 (EET DST) From: Narvi To: "Sexton, Robert" cc: freebsd-hackers@freefall.freebsd.org Subject: Re: Kernel Config (Was: GENERIC Kernel Debate) In-Reply-To: <7366895B0187397C@mg01a.mhs.squared.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, 11 Jul 1996, Sexton, Robert wrote: > > Is it possible for us to get away from the use of compilers in kernel > config? I come from the SCO world, and we have a nice mechanism for > adding and removing devices without recompiling the kernel. It does have > to be relinked, but that's a much smaller job. I'm actually > oversimplifying a little. SCO (And AT&T, I think) use a table-driven > method for deciding what to link in, and then compile a few small > binaries which contain tuning parameters. For turnkey systems, they > supply a small, primitive compiler for the job. > I have touched a few Linux systems, and I'm getting used to the Berkely > style kernel config mechanism. I think it's easier to find source code > on the BSD kernel system, but ease of configuration leaves a lot to be > desired. I don't think make is really a suitable tool for kernel config. > > I think we all ultimately want the all-singing, all-dancing loadable > kernel modules with PnP detection and auto config of devices while > simultaneously selecting proper drivers for compatible hardware while > working around all known incompatibilites. > In the mean time, how about a better way to build a custom kernel, or > tune an existing kernel? > The "right" way would perhaps be a utility which would allow one to specify which drivers, how many instances, etc. one wants to have and compile a kernel config file from that information? Having precompiled drivers, etc. for only kernel linking should be possible by only some digging in the Makefiles - want to have a try? Sander PS. I do understand having precompiled modules for the kernel would not be to everyones taste - but remeber we are not discussing removing the source distribution... > Robert Sexton. > > >