Date: Wed, 25 Feb 1998 23:35:49 -0700 From: "Justin T. Gibbs" <gibbs@plutotech.com> To: "Michael L. VanLoon -- HeadCandy.com" <michaelv@MindBender.serv.net> Cc: Jaye Mathisen <mrcpu@cdsnet.net>, Remy NONNENMACHER <remy@synx.com>, Stephen Ritter <t_sritte@qualcomm.com>, freebsd-smp@FreeBSD.ORG Subject: Re: ASUS P2L97DS Message-ID: <199802260638.XAA19358@pluto.plutotech.com> In-Reply-To: Your message of "Wed, 25 Feb 1998 21:27:21 PST." <199802260527.VAA00950@MindBender.serv.net>
next in thread | previous in thread | raw e-mail | index | archive | help
>[...] >>I still think the FreeBSD way is broke. NT can figure it out, Windows can >>figure it out, Solaris gets it the same way, but then there's FreeBSD out >>on the fringe. Right, but difficult to use. > >Uh, to be honest, this isn't necessarily true. > >I haven't seen this happen with 2940s, but I *have* seen it happen >with multiple NCR53c810 cards. DOS saw them in the order the BIOS >decided they should be probed, which was high address card first, then >the lower one. Actually, the PCI BIOSes may be installed anywhere. The slot probe order, however was probably the problem. Both FreeBSD and NT probe in a hard coded way which may not match the way the BIOS did it. For EISA/ISA/VL cards where you can determine that they installed a BIOS, you can reorder the default bus probe to follow ascending BIOS address which is how 99% of the motherboards in the world scan and execute BIOSes. CAM has provisions for handling a controller defined ordering for attaching buses that does not necessarily match probe order, but it may be better to sit down with one of the PnP books and attempt to make FreeBSD's attach order match, as closely as possible, to what the PnP spec suggests for BIOSes. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802260638.XAA19358>