From owner-freebsd-arch@FreeBSD.ORG Sat Jun 18 12:50:54 2011 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D1298106566B for ; Sat, 18 Jun 2011 12:50:54 +0000 (UTC) (envelope-from freebsd-arch@m.gmane.org) Received: from lo.gmane.org (lo.gmane.org [80.91.229.12]) by mx1.freebsd.org (Postfix) with ESMTP id 8DF718FC0A for ; Sat, 18 Jun 2011 12:50:54 +0000 (UTC) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1QXuki-00013I-Lu for freebsd-arch@freebsd.org; Sat, 18 Jun 2011 14:35:52 +0200 Received: from 42-56.dsl.iskon.hr ([89.164.42.56]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 18 Jun 2011 14:35:52 +0200 Received: from ivoras by 42-56.dsl.iskon.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 18 Jun 2011 14:35:52 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-arch@freebsd.org From: Ivan Voras Date: Sat, 18 Jun 2011 14:35:44 +0200 Lines: 23 Message-ID: References: <20110611171834.GA38142@zim.MIT.EDU> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: 42-56.dsl.iskon.hr User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2.17) Gecko/20110414 Thunderbird/3.1.10 In-Reply-To: <20110611171834.GA38142@zim.MIT.EDU> Subject: Re: [RFC] shipping kernels with default modules? X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Jun 2011 12:50:54 -0000 On 11/06/2011 19:18, David Schultz wrote: > OS X has an interesting solution, intended to preserve the > flexibility of dynamic modules, while minimizing boot time. > It provides a kextcache utility, which packages the kernel > and all of the needed modules into a single binary for better > locality on disk. Unlike recompiling the kernel, running > kextcache is fast, and the system runs it automatically when > hardware or driver changes necessitate it. This seems like the goldilocks solution, but also dangerously close to the Linux's horrendeous initrd system where, if you change the disc controller (which nowadays can mean simply switching it in BIOS from ATA to AHCI to RAID modes) and you get a magically unbootable system. I'd go for something like that only if there's a fallback mode acting like "ok, the previous boot failed, now automagically load ALL the modules and try again". You pretty much need to load all the drivers because except for some obvious exceptions (like the sound drivers, VESA, etc.) many can host devices for file systems ("normal" controllers, network cards, iSCSI, GEOM, USB, Infiniband, etc.). Doing it without this kind of fallback mode is a step backwards.