From owner-freebsd-bugs Fri Mar 17 18:46:27 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id SAA06180 for bugs-outgoing; Fri, 17 Mar 1995 18:46:27 -0800 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id SAA06173 for ; Fri, 17 Mar 1995 18:46:15 -0800 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id MAA05056; Sat, 18 Mar 1995 12:43:56 +1000 Date: Sat, 18 Mar 1995 12:43:56 +1000 From: Bruce Evans Message-Id: <199503180243.MAA05056@godzilla.zeta.org.au> To: bde@zeta.org.au, terry@cs.weber.edu Subject: Re: Can't install 2.0-RELEASE on EIDE HD Cc: bugs@FreeBSD.org, gineoh@engin.umich.edu Sender: bugs-owner@FreeBSD.org Precedence: bulk >> > cylndrs heads sectors size >> > 525 64 63 1083 megabytes >> >> This configuration might work iff the BSD partition starts on the first >> head on the first cylinder, i.e., on a sectore < 63. >Actully, it should have intermittent areas of 16 sectors throughout >the disk where this would work. Actually, you'd have to look at the pathology in the driver to see what it would do... it's closer to working that I thought. The driver modifies the geometry in the label from (525, 64, 63, secperunit = correct, secpercyl = 64*63) to (525, H, 63, secperunit = correct, secpercyl = 64*63) where H = (number of heads reported by drive or 16 if drive couldn't report it). Note that the driver doesn't modify secpercyl, so it screws up the C/H/S calculations later; otherwise the modified geometry would work on modern drives. If H == 16, then cylinder addresses given to the controller will be too small by a factor of 4. Bruce