From owner-freebsd-current Tue Mar 19 23:31:35 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id XAA10553 for current-outgoing; Tue, 19 Mar 1996 23:31:35 -0800 (PST) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id XAA10538 Tue, 19 Mar 1996 23:31:28 -0800 (PST) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id XAA22233; Tue, 19 Mar 1996 23:31:18 -0800 From: "Rodney W. Grimes" Message-Id: <199603200731.XAA22233@GndRsh.aac.dev.com> Subject: Re: aic7xxx driver problem with wide and narrow busses attached. To: gibbs@freefall.freebsd.org (Justin T. Gibbs) Date: Tue, 19 Mar 1996 23:31:17 -0800 (PST) Cc: mrl@teleport.com, current@Freebsd.Org In-Reply-To: <199603200707.XAA08756@freefall.freebsd.org> from "Justin T. Gibbs" at Mar 19, 96 11:07:01 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-current@Freebsd.Org X-Loop: FreeBSD.org Precedence: bulk > > >> > >> > >> Disable parity detection in SCSI-Select. I'm still trying to > >> track down why this happens on machines that have both a > >> Wide and narrow bus attached. (Hey Rod... care to lend me > >> a wide disk again? 8-) > >> > >I'll have to do some house cleaning... my spare wide drive is a bit > >full of data right now: > >/dev/sd1a 2088244 1936934 -15750 101% /B > > > >> I'll turn parity error checking off again in the driver for now. > > > >I'll slap a narrow disk on my 2940W and see if I can duplicate it > >here, if it does duplicate I'll see about getting this data rolled > >out to other disks and get it down to you for a debug session... > > It may only occur on the 7880 based controllers. Do you have any > older 2940Ws around? The only 2940W I have is: ahc0 rev 3 int a irq 11 on pci0:12 ahc0: aic7870 Wide Channel, SCSI Id=7, 16 SCBs I don't currently have any 2940UW's. > >Should I be able to duplicate this in -stable, or does this problem > >only seem to be occuring with -current? > > I only enabled parity checking for the aic78X0 family recently in > -current. I didn't even think much about it since it was an > oversight that it wasn't enabled before and it has been enabled > for some time on the aic7770 chips. The setting tracks what is > set in SCSI-Select so its easy to change. I should have the -current > driver pulled back into -stable soon (minus the parity stuff). I'll > send you a private patch to enable the parity once I do this. Okay, but make sure that commit message to -stable clearly states that if you suddenly see SCSI problems on aic7xxx based systems to turn off parity in the SCSI-Select menu and report the problem. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD