From owner-freebsd-hackers  Sat Nov  1 21:37:06 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id VAA00544
          for hackers-outgoing; Sat, 1 Nov 1997 21:37:06 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from trojanhorse.ml.org (mdean.vip.best.com [206.86.94.101])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA00537
          for <hackers@freebsd.org>; Sat, 1 Nov 1997 21:37:03 -0800 (PST)
          (envelope-from jamil@trojanhorse.ml.org)
Received: from localhost (jamil@localhost)
	by trojanhorse.ml.org (8.8.7/8.8.5) with SMTP id VAA00793;
	Sat, 1 Nov 1997 21:36:30 -0800 (PST)
Date: Sat, 1 Nov 1997 21:36:30 -0800 (PST)
From: "Jamil J. Weatherbee" <jamil@trojanhorse.ml.org>
To: Mike Smith <mike@smith.net.au>
cc: hackers@freebsd.org, dufault@hda.com
Subject: Re: Strategy Routines 
In-Reply-To: <199711020252.NAA00521@word.smith.net.au>
Message-ID: <Pine.BSF.3.96.971101212300.671C-100000@trojanhorse.ml.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


Ok, so say I am doing outb(), one bang after the other. Since timo can
sleep up to 1/100 of a second. What would be a good "packet size" (i.e.
number of outb() to limit myself to in a loop. My initial estimates would
indicate that I could break up the buffer in something like:
  (this is just for example I know the modulus is pretty damn
innefficient)

for (i=0; i<len; buf++)
 {
    outb (port, *buf)
    
    if (!(++i % 2048)) tsleep (myself, PPAUSE, "iosink", 1); 
 }




> This is a fault in the driver design.  Drivers should not hog the 
> cpu, obviously.  Try calling tsleep() out of the driver with a very 
> short timeout.
> 
> mike
> 
>