From owner-freebsd-hardware  Mon Apr  1 10:23:50 1996
Return-Path: owner-hardware
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id KAA12848
          for hardware-outgoing; Mon, 1 Apr 1996 10:23:50 -0800 (PST)
Received: from ref.tfs.com (ref.tfs.com [140.145.254.251])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id KAA12809
          Mon, 1 Apr 1996 10:23:41 -0800 (PST)
Received: (from julian@localhost) by ref.tfs.com (8.7.3/8.6.9) id KAA04704; Mon, 1 Apr 1996 10:21:49 -0800 (PST)
Message-Id: <199604011821.KAA04704@ref.tfs.com>
Subject: Re: Changes to FreeBSD kernel to keep "green" drives on
To: Brett_Glass@ccgate.infoworld.com (Brett Glass)
Date: Mon, 1 Apr 1996 10:21:49 -0800 (PST)
From: "JULIAN Elischer" <julian@ref.tfs.com>
Cc: freebsd-hardware@FreeBSD.org, hackers@FreeBSD.org, bugs@FreeBSD.org
In-Reply-To: <9603018283.AA828378293@ccgate.infoworld.com> from "Brett Glass" at Apr 1, 96 10:00:44 am
X-Mailer: ELM [version 2.4 PL25 ME8b]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hardware@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> 
> Here are the diffs to FreeBSD 2.1.0-RELEASE that turn off the inactivity
> timer. Turning on these flags will always be a good idea for desktop
> systems with "green" hard drives, but the kernel still needs fixing to
> handle systems in which an inactivity timeout is desirable (e.g. laptops).
My laptop has always beed set up to timeout the drive after 30 secons..
the system has been coping wth this successfully since 386BSD0.1 (1992).
the drive takes about 1.5 seconds to spin-up...
the standard timeoutes in the driver seem to cope.. 
when the batery is low it sometimes has to try several times
to overcome 'stiction' (you can hear teh drive kick the spindle a couple
of times before it starts,, the driver seems to handle this as well,
though it takes about 4 seconds to get going in this case.

I know phk has been looking at spin-up optimisatiosn..
(e.g. if a drive takes more than a set time to  come ready, assumen it was
spinning up and schedule a 'sync'  to catch it while it's already going..