From owner-cvs-all Wed Sep 22 21:43:47 1999 Delivered-To: cvs-all@freebsd.org Received: from overcee.netplex.com.au (overcee.netplex.com.au [202.12.86.7]) by hub.freebsd.org (Postfix) with ESMTP id 098BD15008; Wed, 22 Sep 1999 21:43:40 -0700 (PDT) (envelope-from peter@netplex.com.au) Received: from netplex.com.au (localhost [127.0.0.1]) by overcee.netplex.com.au (Postfix) with ESMTP id 633D01CA7; Thu, 23 Sep 1999 12:43:35 +0800 (WST) (envelope-from peter@netplex.com.au) X-Mailer: exmh version 2.0.2 2/24/98 To: nate@mt.sri.com (Nate Williams) Cc: Greg Lehey , Matt Dillon , cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: Building klds (was: cvs commit: src/sys/i386/include param.h src/sys/alpha/include param.h src/sys/conf options src/sys/i386/conf LINT) In-reply-to: Your message of "Wed, 22 Sep 1999 18:41:44 CST." <199909230041.SAA17174@mt.sri.com> Date: Thu, 23 Sep 1999 12:43:35 +0800 From: Peter Wemm Message-Id: <19990923044335.633D01CA7@overcee.netplex.com.au> Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk Nate Williams wrote: > > > Modified files: > > > sys/i386/include param.h > > > sys/alpha/include param.h > > > sys/conf options > > > sys/i386/conf LINT > > > Log: > > > Back out a portion of the last commit. DFLTPHYS and MAXPHYS cannot > > > be set by a kernel conf option due to the struct buf structural > > > dependancy (sizing of b_pages[]) creating a conflict with modules > > > (which are not compiled with kernel config options overrides). > > > > This really points out a basic deficiency of the current build > > process: klds should be built along with the kernel, not along with > > userland, and they should have access to the kernel config files. > > One of the (many) goals of KLD's is that you can compile them one system > and transport them to another. Your config file is way different from > mine, so this is a losing strategy. Not necessarily. KLD's intended for distribution should be built with a GENERIC-style configuration so they support all the variations of cpu type, etc. If you choose to change the local core kernel services and interfaces to be incompatable with distributed KLD's, then we should support that as it's a necessary part of development and local tuning. For example, it should be easy to build a coherent kernel+modules that use a different MAXPHYS and hence struct buf. The anti-foot-shooting measures are going to be interesting though. Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message