[prev in list] [next in list] [prev in thread] [next in thread] 

List:       uclinux-dev
Subject:    RE: [uClinux-dev] Setting ethernet MAC address at productiontime
From:       "Paul Carpenter" <paul.carpenter () eng ! nciaccess ! com>
Date:       2002-01-29 14:22:41
[Download RAW message or body]

> > I was about to ask the very same question.  Also, I am confused
> about the
> > M5272C3's handling of the MAC address.  It would appear (from
> source code)
> > that dBUG doesn't actually write the new MAC address to the CPU's SIM
> > registers when you do a "set mac <value>".  It only writes the
> new value to
> > Flash, then retrieves the value from Flash parm's section when
> you "get" it.
> > In the uClinux kernel, in fec_init(), there doesn't appear to
> be any code
> > that sets up the MAC either.
>
> That is right. The driver assumes that dBUG has already set
> the MAC address registers. This is only true if you do a network
> download through dBUG.
>
> >  Therefore, I am assuming that the MAC reported
> > by the board (when you run "ifconfig") is actually the default value at
> > which the SIM powers up.  Is this correct?  BTW, we've been
> getting a value
> > of 00:d0:cf:00:00:72 from "ifconfig".
>
> Which date/version of uClinux are you using?
> This was a bug in an older version. The FEC driver was hard
> coding the MAC address to 00:d0:cf:00:00:72.

We are using 2.0.38.1pre7.  You are correct: there was a bug in which the
FEC driver hard-coded that MAC address.  I've changed it such that it now
reads the MAC address directly from the SIM if the MCF5272 is present, and
modified our dBUG such that the desired address (stored in Flash) is always
written out to the SIM.


Paul Carpenter (Firmware Engineer)
Northern Computers, Inc.
Milwaukee, WI
414.769.5980 x4322 (voice)
414.769.5989 (fax)

This message resent by the uclinux-dev@uclinux.org list server http://www.uClinux.org/
[prev in list] [next in list] [prev in thread] [next in thread]