[ moved to port-sparc, as it appears to be a sparc-specific problem ]
   
   
   i did actually pin point the date (within a couple of days) that
   the lossage occured, but i've forgotten it no.  :-(  i think it
   might have been around 3rd december, or 3rd january (the `3rd'
   part is what i remember, i think.  :-)

ah, here it is.  this is a reply to my first query about this problem
to port-sparc:

-- 
From: Chuck Silvers <chs@wbs.net>
To: port-sparc@NetBSD.ORG
Subject: Re: current kernel busted zs driver?
Date: Wed, 20 Nov 1996 12:05:07 -0800

I've been having the same trouble from -current from a week or two ago.
It doesn't always get the fifo overrun, just most of the time.
The most recent kernel I have that doesn't have this problem is from Nov. 3.

-Chuck
-- 


i actually went and diffed the entire kernel tree between november
2 and november 19 (when i noticed the problem first!), and didn't
see anything that looked likely.  :-(


.mrg.