From owner-freebsd-hardware Mon Aug 5 02:43:30 1996 Return-Path: owner-hardware Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id CAA05180 for hardware-outgoing; Mon, 5 Aug 1996 02:43:30 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id CAA05166 for ; Mon, 5 Aug 1996 02:43:01 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id TAA09932; Mon, 5 Aug 1996 19:37:04 +1000 Date: Mon, 5 Aug 1996 19:37:04 +1000 From: Bruce Evans Message-Id: <199608050937.TAA09932@godzilla.zeta.org.au> To: BRETT_GLASS@ccgate.infoworld.com, phk@critter.tfs.com Subject: Re: Mapped geometry vs. Actual Cc: hardware@freebsd.org, msmith@atrad.adelaide.edu.au, randy@zyzzyva.com, rgrimes@GndRsh.aac.dev.com Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >>>> It seems as if it also tries to force disk slices to >>>> cylinder boundaries -- which is silly on ZBR drives or when sector >>>> mapping is done. >>... >>I said "slices," not "partitions." FreeBSD seems to insist upon moving >>things to cylinder boundaries even within partitions. >That is also for backwards compatibility mode. 386BSD :-) NetBSD, OpenBSD, ... I don't think there was ever such a restriction. However, ufs wants to do things in terms of cylinder groups so partial cylinder [groups?] at best wastes space for partitions that contain ufs file systems. It sometimes wastes a lot of space - try newfs with the default parameters on a floppy. Bruce