From owner-freebsd-hackers  Fri Dec 15 12:37:23 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id MAA16287
          for hackers-outgoing; Fri, 15 Dec 1995 12:37:23 -0800 (PST)
Received: from ns.ge.com (ns.ge.com [192.35.39.24])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA16282
          for <FreeBSD-hackers@FreeBSD.org>; Fri, 15 Dec 1995 12:37:08 -0800 (PST)
Received: from crissy.gemis.ge.com ([3.29.7.57]) by ns.ge.com (8.6.12/8.6.11) with ESMTP id PAA18091; Fri, 15 Dec 1995 15:36:54 -0500
Received: from salem.ge.com (carsdb.salem.ge.com [3.29.7.15]) by crissy.gemis.ge.com (8.6.11/8.6.11) with SMTP id PAA20620; Fri, 15 Dec 1995 15:30:59 -0500
Received: from combs.salem.ge.com by salem.ge.com (4.1/SMI-4.1)
	id AA17483; Fri, 15 Dec 95 15:36:49 EST
Received: (from steve@localhost) by combs.salem.ge.com (8.7.2/8.6.11) id PAA26801; Fri, 15 Dec 1995 15:36:49 -0500 (EST)
Date: Fri, 15 Dec 1995 15:36:49 -0500 (EST)
From: "Stephen F. Combs" <combssf@salem.ge.com>
Message-Id: <199512152036.PAA26801@combs.salem.ge.com>
To: bde@zeta.org.au, nate@rocky.sri.MT.net
Subject: Re: sio overruns on 2.0.5R
Cc: FreeBSD-hackers@FreeBSD.org
Sender: owner-hackers@FreeBSD.org
Precedence: bulk


> From owner-freebsd-hackers@freefall.freebsd.org Fri Dec 15 15:29:45 1995
> Date: Fri, 15 Dec 1995 12:06:07 -0700
> From: Nate Williams <nate@rocky.sri.MT.net>
> To: Bruce Evans <bde@zeta.org.au>
> Cc: nate@rocky.sri.MT.net, FreeBSD-hackers@FreeBSD.org
> Subject: Re: sio overruns on 2.0.5R
> In-Reply-To: <199512151859.FAA26342@godzilla.zeta.org.au>
> Sender: owner-hackers@FreeBSD.org
> Content-Length: 841
> 
> > >> After I switched to a Asus P55TP4XE 90Mc I see things like
> > >> 
> > >> sio0: 5 more silo overflows (total 7)
> > 
> > >I hate to sound like a broken record, but are you *sure* you're using
> > >flow-control?  Can you use stty on the connected end to make sure?
> > 
> > Er, flow control has no effect on silo overflows.
> 
> Really?  If a machine doesn't provide flow control can't you over-run
> the buffers if the machine gets loaded down and the interrupt load gets
> so bad that the sio interrupt handler isn't called enough.  However, if
> you use flow control shouldn't it stop the buffer from getting
> overflowed simply because no more data is being received.
> 
> Am I totally out to lunch here?  I know that I get lots of silo
> over-flows when I don't use flow control and push the line fast, where
> as soon as I turn on flow control they go away.
> 
> 
> Nate
> 
Unless the FreeBSD team responsible for the 'sio' driver has made ENORMOUS
changes since the last time I looked at it (quite a while ago!?!),
flow control DOES affect silo-overruns!  Flow-Control provides a throttle, 
limiting the chances of a silo-overrun (not eliminating it, just limiting it).
The modem/printer/mouse/other computer attached to the sio port MUST respect
the STOP flow-control signal within a reasonable amount of time
(a few, 2-4, character transmission times, usually) for flow-control to be
effective, but, most modern pieces of equipment do.

Steve Combs