From owner-freebsd-current  Tue Apr  1 11:10:51 1997
Return-Path: <owner-current>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA21356
          for current-outgoing; Tue, 1 Apr 1997 11:10:51 -0800 (PST)
Received: from dyslexic.phoenix.net (dyslexic.phoenix.net [199.3.233.7])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA21341
          for <current@freebsd.org>; Tue, 1 Apr 1997 11:10:38 -0800 (PST)
Received: (from gemohler@localhost) by dyslexic.phoenix.net (8.8.5/8.7.3) id NAA11158; Tue, 1 Apr 1997 13:09:00 -0600 (CST)
Date: Tue, 1 Apr 1997 13:09:00 -0600 (CST)
From: Geoff Mohler <gemohler@c-com.net>
X-Sender: gemohler@dyslexic.phoenix.net
To: "John S. Dyson" <toor@dyson.iquest.net>
cc: current@freebsd.org
Subject: Re: Bounce Buffers
In-Reply-To: <199704011752.MAA02576@dyson.iquest.net>
Message-ID: <Pine.BSF.3.91.970401130747.10066A-100000@dyslexic.phoenix.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Tue, 1 Apr 1997, John S. Dyson wrote:

> > I was getting this on a stock 2.2.1 installation:
> > 
> > Panic: Bounce Buffers Out Of Range
> > =-=-=-=-=-
> > 
> > I got this when I took the GENERIC *or any* Kernel definition file, and 
> > added the line:
> > 		options		"MAXMEM=262144"
> > 
> > Then it would report the above error on bootup, just after reporting 
> > kernel bus options.
> > 
> > Now, the problem only went away after removing a megabyte of ram.  When I 
> > changed it to:
> > 		options		"MAXMEM=261120"
> > 
> > ..it works fine.
> > 
> You can also remove the bounce-buffers define.  Those aren't usually needed
> with new technology hardware.

Yes, even with and without the:	options		BOUNCE_BUFFERS
...it still did the same thing.

Note:  Worked with Rod Grimes on this issue.
 
> The problem is that the kernel takes so much space, that there isn't room
> <16MB for bounce buffers.  Unless we get lots of users with > 128M using
> ISA SCSI adaptors, I don't think that it is worth fixing the memory
> allocation code, until we create physical memory type attributes.

Interesting note..but this still happens with and without the 
bounce_buffers option.

=-=-=-=-=-
ROMANI ITE DOMUM

 Geoff Mohler                                               
 Operations Engineer       
 Charter Communications/Phoenix Data Net