From owner-freebsd-hackers  Fri May 31 09:45:23 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id JAA22951
          for hackers-outgoing; Fri, 31 May 1996 09:45:23 -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 JAA22944
          for <freebsd-hackers@freefall.freebsd.org>; Fri, 31 May 1996 09:45:19 -0700 (PDT)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id CAA12346; Sat, 1 Jun 1996 02:39:59 +1000
Date: Sat, 1 Jun 1996 02:39:59 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199605311639.CAA12346@godzilla.zeta.org.au>
To: bde@zeta.org.au, freebsd-hackers@freefall.freebsd.org, mrm@MARMOT.Mole.ORG,
        sextonr.crestvie@squared.com
Subject: Re: Disktab and SCSI Performance.
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>Rotdelay and/or artificially increased interleave can be useful for
>fast disks, too. It can be used to prevent a fast drive on a non-dma
>controller from starving the CPU to the exclusion of user programs.
>Sometimes slowing disk I/O can be an advantage. From experience :-)

Can it be used to prevent fast ID drives from starving the CPU to the
exclusion of kernel code? :-)

Bruce