From owner-freebsd-hackers Fri Dec 28 14:31:50 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from mcqueen.wolfsburg.de (pns.wobline.de [212.68.68.5]) by hub.freebsd.org (Postfix) with ESMTP id 34F3F37B41B; Fri, 28 Dec 2001 14:31:44 -0800 (PST) Received: from colt.ncptiddische.net (ppp-294.wobline.de [212.68.71.15]) by mcqueen.wolfsburg.de (8.11.3/8.11.3/tw-20010821) with ESMTP id fBSMUk805099; Fri, 28 Dec 2001 23:30:46 +0100 Received: from tisys.org (poison.ncptiddische.net [192.168.0.5]) by colt.ncptiddische.net (8.11.6/8.11.6) with ESMTP id fBSMUoX81871; Fri, 28 Dec 2001 23:30:51 +0100 (CET) (envelope-from nils@tisys.org) Received: (from nils@localhost) by tisys.org (8.11.6/8.11.6) id fBSMUix04563; Fri, 28 Dec 2001 23:30:44 +0100 (CET) (envelope-from nils) Date: Fri, 28 Dec 2001 23:30:09 +0100 From: Nils Holland To: =?iso-8859-1?Q?S=F8ren_Schmidt?= Cc: Glenn Johnson , Greg Lehey , FreeBSD Questions , FreeBSD Hackers Subject: Re: Does FreeBSD have a problem with some AMD processors? Message-ID: <20011228233009.A4519@tisys.org> References: <20011228204612.GA1226@gforce.johnson.home> <200112282212.fBSMCea36745@freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.2.5i In-Reply-To: <200112282212.fBSMCea36745@freebsd.dk>; from sos@freebsd.dk on Fri, Dec 28, 2001 at 11:12:40PM +0100 X-Operating-System: FreeBSD poison.ncptiddische.net 4.5-PRERELEASE FreeBSD 4.5-PRERELEASE X-Machine-Uptime: 11:19PM up 8:50, 1 user, load averages: 0.02, 0.09, 0.07 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 On Fri, Dec 28, 2001 at 11:12:40PM +0100, Søren Schmidt stood up and spoke: > > >From what I can tell the newest Epox BIOS for the 8kta3 does *not* > install the 686b fix when it doesn't detect a SBLive! sound card, > nice but not good enough :/ There are many information on the 686B bug available on the net, and looking at all of them and trying to draw some conclusions isn't going to work, as much of the information that can be found is conflicting. Actually, I just read an article that "revisited" the 686B in August, a few months after the fix was released. They claim that they were *only* able to produce the problem with a SBLive, but not without. Other reports, and that is the most widely adopted belief, is that the bug can also occur without the SBLive. Other reports I read claim to have found similar problems in the MVP3 chipset, as used in a lot of K6-2 systems. According to these articles, "the 686B is only the latest member of a family of flawed chipsets". More or less randomly, I got about a mailing list post where multiple Linux users reported data corruption with the MVP3. But back to the 686B: I don't know what Linux is currently doing about the bug, but for some time after the problem was discovered, it seems as if they simply turned DMA off (a miling list post on this topic said: "...and now all these high-performance Athlon systems are going to run in 1992 PIO mode by default"). Well, I guess all of this may well be a little confusing ;-) Greetings Nils -- Nils Holland Ti Systems - FreeBSD in Tiddische, Germany http://www.tisys.org * nils@tisys.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message