From owner-freebsd-questions Sun Mar 31 11:37:07 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA17816 for questions-outgoing; Sun, 31 Mar 1996 11:37:07 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id LAA17803 Sun, 31 Mar 1996 11:37:03 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id MAA11535; Sun, 31 Mar 1996 12:32:40 -0700 From: Terry Lambert Message-Id: <199603311932.MAA11535@phaeton.artisoft.com> Subject: Re: Dell EIDE drive data corruption with FreeBSD? To: macgyver@infinet.com (Wilson MacGyver) Date: Sun, 31 Mar 1996 12:32:39 -0700 (MST) Cc: terry@lambert.org, macgyver@infinet.com, hackers@FreeBSD.org, questions@FreeBSD.org In-Reply-To: <315DC137.84D@cylatech.com> from "Wilson MacGyver" at Mar 30, 96 06:18:15 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-questions@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > > Is this the flawed IDE chipset that loses data if you interleave I/O? > > > > If so, the answer is to change your CMOS settings. > > Yes it is, but I thought the FreeBSD IDE driver is not affected by this bug? It does not initiate DMA's. DMA completion is not the only thing that could cause an interrupt. A bad spot on the disk, two disks on the same controller, or an IDE CDROM can also trigger the problem. Since FreeBSD doesn't use bus mastering DMA, it doesn't trigger the bug if all you have is one disk, or two disks from the same manufacturer which are configured correctly (I have heard som reports on Conner, but nothing verified). Either way, it can't *hurt* FreeBSD for you to set the CMOS settings to get around the bug. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.