From owner-freebsd-hackers Sun Sep 9 19:10:20 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from server.soekris.com (soekris.com [216.15.61.44]) by hub.freebsd.org (Postfix) with ESMTP id 2DC1837B407; Sun, 9 Sep 2001 19:10:17 -0700 (PDT) Received: from soekris.com (1.4.soekris.com [192.168.1.4] (may be forged)) by server.soekris.com (8.9.2/8.9.2) with ESMTP id TAA13231; Sun, 9 Sep 2001 19:10:20 -0700 (PDT) (envelope-from soren@soekris.com) Message-ID: <3B9C2102.4AC887E1@soekris.com> Date: Sun, 09 Sep 2001 19:10:10 -0700 From: Soren Kristensen Organization: Soekris Engineering X-Mailer: Mozilla 4.77 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Bsdguru@aol.com Cc: msmith@FreeBSD.ORG, hackers@FreeBSD.ORG Subject: Re: PCI probe reordering? References: <6f.1a7105ea.28cd5bb7@aol.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Hi, Aren't operating systems supposed to use the PCI BIOS calls for finding devices ? So the scanning order would then be controlled by the BIOS, and then always follow the intent of the hardware manufacturer ? Or is there problems with their intent ? Regards, Soren Bsdguru@aol.com wrote: > > In a message dated 9/9/01 5:21:51 PM Eastern Daylight Time, > msmith@freebsd.org writes: > > > The entire PCI probe/attach process is going to have to change, however I > > don't see it changing all that much. It is hard to detect devices that > > are "onboard" as opposed to being in physical slots, and there is no > > consistent mechanism that one can take to always get them "first". > > > > In short, what you're asking for is difficult to achieve consistently, > > and not really worth the effort. Your onboard ethernet is fxp1. Deal > > with it. > > > > "Worth it" depends on what you care about, I suppose. But a commercial > vendor, using freebsd as a platform, will "care" if one motherboard scans one > way and another scans differently, as you cant tell customers "well, if you > have this product the ports are this way, and if you have that product the > ports are another way". While there may not be a generic way to always do it > the way one expects, it doesnt seem unreasonable to have a mechanism to > override the default probe that isnt a nightmare to implement. > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-hackers" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message