From owner-freebsd-hackers Wed Aug 23 23:22:34 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id XAA17387 for hackers-outgoing; Wed, 23 Aug 1995 23:22:34 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id XAA17378 for ; Wed, 23 Aug 1995 23:22:27 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id QAA19754; Thu, 24 Aug 1995 16:18:09 +1000 Date: Thu, 24 Aug 1995 16:18:09 +1000 From: Bruce Evans Message-Id: <199508240618.QAA19754@godzilla.zeta.org.au> To: FreeBSD-hackers@freebsd.org, wilko@yedi.iaf.nl Subject: Re: On ESDI install. Sender: hackers-owner@freebsd.org Precedence: bulk >Related question, might be a silly one: I just noticed that after >installing 205R on the said WD1007 system the disklabel has >rpm=0 and rotdelay=0. Why?? Bugs in sysinstall in 2.0.5. I think you mean interleave, not rotdelay. Interleave is what sysinstall gets wrong. >Rotdelay=0 is used for (e.g.) SCSI drives with caches, yes? rpm and interleave aren't used (except to barf). rotdelay is set by newfs and tunefs, not by sysinstall, fdisk or disklabel. The default is 0. 0 is suitable for all drives with read and write caches, i.e., for modern drives, but can give very bad results if the caching is deficient. My ESDI drives had read caches but not write caches, so they wanted rotdelay=0 for read and rotdelay>0 for write. All nonzero rotdelays tend to be equivalent because they force sequential blocks to be separated on the disk and very few drives are slow enough to require a rotdelay of more than one block time. Bruce