From owner-freebsd-hackers@FreeBSD.ORG Sun Feb 19 09:20:40 2012 Return-Path: Delivered-To: hackers@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 768FA1065672; Sun, 19 Feb 2012 09:20:40 +0000 (UTC) (envelope-from perryh@pluto.rain.com) Received: from agora.rdrop.com (unknown [IPv6:2607:f678:1010::34]) by mx1.freebsd.org (Postfix) with ESMTP id 2DE428FC08; Sun, 19 Feb 2012 09:20:40 +0000 (UTC) Received: from agora.rdrop.com (66@localhost [127.0.0.1]) by agora.rdrop.com (8.13.1/8.12.7) with ESMTP id q1J9KdKZ055593 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Sun, 19 Feb 2012 01:20:39 -0800 (PST) (envelope-from perryh@pluto.rain.com) Received: (from uucp@localhost) by agora.rdrop.com (8.13.1/8.14.2/Submit) with UUCP id q1J9KdnM055592; Sun, 19 Feb 2012 01:20:39 -0800 (PST) (envelope-from perryh@pluto.rain.com) Received: from fbsd81 ([192.168.200.81]) by pluto.rain.com (4.1/SMI-4.1-pluto-M2060407) id AA08578; Sun, 19 Feb 12 01:15:00 PST Date: Sun, 19 Feb 2012 08:13:48 -0800 From: perryh@pluto.rain.com To: dougb@freebsd.org Message-Id: <4f411fbc.5xpQwqtOGVzi8G4D%perryh@pluto.rain.com> References: <4F3E8225.9030501@FreeBSD.org> <4F3E8C26.3080900@FreeBSD.org> <4F3EA5F2.9070804@gmail.com> <4F3EAE5F.6070903@gmail.com> <20120217.220802.988.2@DOMY-PC> <4F3EDEBC.7040703@gmail.com> <4F3EFB70.5000102@FreeBSD.org> <4f3ff151.FznGzC6RC0a5qBKx%perryh@pluto.rain.com> <4F403C5E.4000104@FreeBSD.org> In-Reply-To: <4F403C5E.4000104@FreeBSD.org> User-Agent: nail 11.25 7/29/05 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: sendtomatt@gmail.com, rank1seeker@gmail.com, hackers@freebsd.org Subject: Re: 8 to 9: Kernel modularization -- did it change? X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 19 Feb 2012 09:20:40 -0000 Doug Barton wrote: > On 02/18/2012 10:43, perryh@pluto.rain.com wrote: > > Doug Barton wrote: > >> loading modules through loader.conf is > >> veeeeeerrrrryyyyyy sssssllllloooooowwwwww ... > > > > Is it noticeably slower to load (say) a 6MB kernel + 2MB of > > modules than to load an 8MB kernel? > > I don't know, that wasn't the problem I was trying to solve. Given the context of the thread, this: > >> loading modules through loader.conf is > >> veeeeeerrrrryyyyyy sssssllllloooooowwwwww ... seemed to be an objection to modularizing the kernel. Hence my question: is it in fact noticeably slower to load a minimal kernel plus needed modules than to load a kernel that had all those modules built in? Based on the below, I think we agree that the answer is likely to be no, even if all the modules in question were loaded via loader.conf (and the modular version might well load noticeably _faster_ if a sizeable fraction of the modules could be loaded via kld_list instead). > If your question is, "6 + 2-in-loader-conf" then I imagine that > it would be about the same speed, maybe a little slower due to > extra file open-read-close cycles. If it's "6 + 2-in-kld_list" > then I imagine it would be quite a bit faster than an 8 M kernel > ... That is what I would expect, also. > but I look forward to the results of your testing. :) You're asking me to test _your_ assertion? I had expected that you would already have the data to back it up.