From owner-freebsd-arch@FreeBSD.ORG Thu Feb 21 10:07:09 2008 Return-Path: Delivered-To: arch@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0B6D116A402 for ; Thu, 21 Feb 2008 10:07:09 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.freebsd.org (Postfix) with ESMTP id DC15613C469 for ; Thu, 21 Feb 2008 10:07:08 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 611F346BAB; Thu, 21 Feb 2008 05:07:08 -0500 (EST) Date: Thu, 21 Feb 2008 10:07:08 +0000 (GMT) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= In-Reply-To: <86ablvuzgx.fsf@ds4.des.no> Message-ID: <20080221100156.V52922@fledge.watson.org> References: <86odacc04t.fsf@ds4.des.no> <20080219233217.GS27248@funkthat.com> <20080220111157.H44565@fledge.watson.org> <86ablvuzgx.fsf@ds4.des.no> MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="621616949-1809286258-1203588428=:52922" Cc: arch@freebsd.org, John-Mark Gurney Subject: Re: dev.* analogue for interfaces 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: Thu, 21 Feb 2008 10:07:09 -0000 This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. --621616949-1809286258-1203588428=:52922 Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE On Wed, 20 Feb 2008, Dag-Erling Sm=F8rgrav wrote: > Robert Watson writes: > >> We also support interface renaming... Does newbus mind if you rename th= e=20 >> devices in its tree? > > I'm not sure whether you're replying to my proposal or to Julian's=20 > interpretation / extrapolation of it... but I have no intention of hooki= ng=20 > interfaces into newbus. I just want a sysctl tree for struct ifnet like = we=20 > have a sysctl tree for device_t, to access interface parameters which are= =20 > not easily accessible through ifconfig. Hmm. When I look at net/if.c, I don't see renaming support, so perhaps thi= s=20 was just a proposal I was thinking of and not actual code. In either case,= I=20 think the question stands: in a world where interface renaming is supported= ,=20 is your plan to also rename the if.X sysctl tree created for the interface?= =20 Does sysctl have a facility to do this? I assume that somehow the details of your plan involve automatically creati= ng=20 a root node for the interface in if_attach and then exposing the node to th= e=20 driver, possibly via a new pointer in struct ifnet? I'm certainly fine wit= h=20 such a notion, but think we should establish, for devices with a number of= =20 sysctl trees (i.e., dev.em vs if.em, dev.da0 vs disk.da0, etc), a general= =20 philosophy for placing nodes in one or the other somewhat deterministically= =2E Robert N M Watson Computer Laboratory University of Cambridge --621616949-1809286258-1203588428=:52922--