In message <3150AC87-4156-4AE2-90D2-EEDB4E436EB6@shagadelic.org>, Jason Thorpe 
writes:
>
>On Sep 11, 2005, at 10:39 PM, Allen Briggs wrote:
>
>> There is a bug, but it's not a missing PHY.  It detects the media--I
>> presume that's done correctly.  But it's not quite doing the right
>> thing.  I've heard reports from at least one other person that note
>> the same behavior.
>
>Indeed.  Note that for a long time, I used an 82543 card with Fiber.   
>I think I still even have that card in a box of spares...  Anyway, it  
>worked fine for me when I was using it.
>
My wm problem -- as I posted a few days ago, it won't autonegotiate 
10baseT -- did not occur in 2.0; it does in 3.0beta and -current.

Your statement is not inconsistent with what we're seeing...

		--Steven M. Bellovin, http://www.cs.columbia.edu/~smb