Subject: Re: 1.0 Binary XFree86-3.1 Keyboard Problem?
To: None <tmm@databook.com, mycroft@ai.mit.edu>
From: Terry Moore <tmm@databook.com>
List: port-i386
Date: 02/10/1995 07:39:54
>    Perhaps someone with more information can comment on the patch
>    level of the XF86 binary portion of the Walnut Creek CDROM.
> 
> I, for one, wasn't even aware that WC was distributing NetBSD.  Given
> certain statements by WC employees in the past, this frankly surprises
> the Hell out of me, and I would personally recommend against buying a
> CD from them.
> 
> Is it the complete 1.0 distribution, including the binaries for all
> the architectures they were made for (i386, sparc, amiga, mac, and
> pc532, I think)?

Seems to be.  Everything seems to be here.  I've been working with it
for a couple of weeks, and everything seems to be quite complete.  Done
kernel builds and everything.  I really cannot complain.  The only 
semi-bummer about the WC CD is that most of the 'ports' are targeted
for FreeBSD.  Whoever hacked up the openwin package added lots of 
#ifndef __FreeBSD__...  it was easy to find and fix, but still
moderately ugly.  

I want to say what a pleasure the NetBSD experience has been
so far.  This is just like downtown.

By the way, I earlier commented that I was having trouble getting PPPD
to hang up the phone (maybe that was on port-i386, so you may not have seen
it).  In any event, I found the problem:  if the option "proxyarp" is
requested, pppd hangs when it tries to exit.  I have not diagnosed the
problem further than to find the call to get_ether_addr() in sys-bsd.c,
called by sifproxyarp(), also in sys-bsd.c.   Evidently it's only getting
called on shutdown.

My hardware:
	486DX100, NE-2000 (taiwan clone w/16K) as ed2, 16550 for the ppp
	link.  I mention the hardware only because it appears as if
	the hang is in the loop doing the ioctl() -- but that may be
	a red herring.

Don't need proxyarp, so I ain't gonna trouble my mind any more.

--Terry Moore
Databook Inc.