-----Original Message-----
From: Jason Thorpe <thorpej@nas.nasa.gov>
To: Jason Wright <jason@thought.net>
Cc: Paul Kranenburg <pk@cs.few.eur.nl>; Simon Raahauge DeSantis
<xiamin@ghostpriest.rakis.net>; port-sparc@netbsd.org
<port-sparc@netbsd.org>
Date: Wednesday, January 27, 1999 3:47 PM
Subject: Re: qec/be


>On Wed, 27 Jan 1999 14:01:15 -0500 (EST)
> Jason Wright <jason@thought.net> wrote:
>
> > The 'hme' and 'qec' fall into the category of needing to manually frob a
> > control register based on the status of duplex in the PHY.
>
>Is the "qec" the MAC?!  I thought it was the DMA engine/interrupt glue.
>
Oops, yes, bad use of nomenclature on my part.  The 'qec' is exactly as you
say (interrupt/dma glue)... It's the 'be' child of the 'qec' that must be
twiddled based on the phy.

>It may certainly be the case that no MACs that come connected to "qec"s
>support full-duplex operation.
>
Could be... Anyone with a qec+be, an external phy, and an etherswitch wanna
give it a shot?

--Jason Wright
  jason@thought.net
  jason@openbsd.org