Subject: Re: a thought about FFS parameters & disk performance
To: David Jones <dej@inode.org>
From: Curt Sampson <cjs@portal.ca>
List: current-users
Date: 03/30/1997 08:42:50
On Sun, 30 Mar 1997, David Jones wrote:

> - Interleave is probably meaningless.

Yes. As is maxcontig/rotdelay. The rotdelay default was changed
from 2 to 0 between 1.0 and 1.1 (I believe) because that's a better
default for modern disks.

> IMHO, with today's disks, a LOT of the "optimizations" that were quite valid
> in the days of the Vax 11/780 are useless today, and the FFS code can do
> with a few simplifications as a result.

I've thought about this a couple of times, and I've come to the
conclusion that doing these simplifications would be a bad idea.

First, it's changing code that has been extensively tested and is
fairly bug-free at this point for no real gain.

Second, there are systems out there that NetBSD is running on, such
as the VAX port, where the extra code still provides a performance
advantage.

cjs

Curt Sampson    cjs@portal.ca		Info at http://www.portal.ca/
Internet Portal Services, Inc.		`And malt does more than Milton can
Vancouver, BC   (604) 257-9400		 To justify God's ways to man.'