From owner-freebsd-hackers  Sun Jan 28 14:23:06 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id OAA05528
          for hackers-outgoing; Sun, 28 Jan 1996 14:23:06 -0800 (PST)
Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id OAA05510
          for <hackers@FreeBSD.org>; Sun, 28 Jan 1996 14:22:58 -0800 (PST)
Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA01704; Sun, 28 Jan 1996 15:19:21 -0700
From: Terry Lambert <terry@lambert.org>
Message-Id: <199601282219.PAA01704@phaeton.artisoft.com>
Subject: Re: POSIX.4 scheduler interface for FreeBSD-2.1
To: dufault@hda.com (Peter Dufault)
Date: Sun, 28 Jan 1996 15:19:21 -0700 (MST)
Cc: jau@jau.csc.fi, hackers@FreeBSD.org
In-Reply-To: <199601281510.KAA13795@hda.com> from "Peter Dufault" at Jan 28, 96 10:10:42 am
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hackers@FreeBSD.org
Precedence: bulk

> (...)
> 
> Thanks Jukka.
> 
> On a related note, can anyone think of anything that would break
> if all drivers used start queues that were priority queues based
> on real time priority instead of FIFO?
> 
> Unrelated I/O would finish out of order but I can't see why that
> would matter.

Windows 95 reenables interrupts immediately after queueing a descriptor
for servicing.  It seems that their code has much less effective
latency for interleaved events because of this.

This type of approach would lend itself to this type of driver
capability.


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.