I have been running NetBSD on a machine for over a year now and 
now I am running it on a second machine.  I have been following the
list in archive for some time, but I have just only subscribed. 

Anyway, that having been said, I have a few questions:

1: I would like to use one of the machines to hook up to various HP-IB
(IEEE 488) instruments around the lab as needed, in particular a 
LeCroy Oscilloscope.  Now, I don't expect the kernel to understand 
every instrument that I come across, but what I do want is a "generic
HP-IB" device similar to Linux's generic SCSI device that would allow 
a user level program to interact with a particular device on the
chain.  Ideally I could open up "/dev/hpib0.3" and be able to
read/write to a the 3rd device on the 0th interface and perhaps 
use ioctl's to do the low-level stuff like reset the device or
whatever.  Does such a device exist?  I don't seem to have an hpib(4)
man page and I haven't gotten deep enough into picking apart the
source code to see if this is implemented. 

2: I would like to compile a stripped down kernel (without the devices
that I don't have between the 2 computers).  I have a obtained the aug
3rd snapshot from gatekeeper.dec.com:/....../sys.tar.gz, but config
complains of no files.hp300 file.  I am much more familiar with Linux,
but it doesn't run on this hardware (yet).

3: I have the X11R5 server and it seems to work for most things, but
I can crash it at *any* time just by running xfig.  reliably, every
time, from local or remote, doesn't matter.  Is this a known bug? 
There was a lot of talk back in May about getting an X11 server
together, is anything happening toward that? 

4: I grabbed the X11R6 stuff, but it seems to need a lot of extra libs to
compile clients and because of this the load time for the clients is 
really slow.  I have dropped back to the X11R5 stuff for now. 

5: One of the machines has the BNC ethernet and it kicks out a lot 
of "le0: CRC" and "le0: FRAM" over time.  I don't know if this is 
a flaky cable or hardware or just the net over-loaded, but i have 
just ignored this till now because it basically works. 

The machines in question are hp340's with 16 MHz 68030's with 68882's, 
Both have 16 Meg of ram, dca0.  One has a fast-hpib and 3 7959B's 
and the other only a normal hpib and 1 7959B.  The one with the fast
HPIB has a 1024 x 768 64 color lo-res catseye display, and the other 
a 1024 x 768 monochrome topcat display.   Oh, and when xfig kills X,
it wipes out the entire machine on the catseye display (funny snow on
the monitor then a hard, unhealthy hang), but only dumps core on the 
topcat.  Both have ROM version C1. 


Anyway, I think i have written enough for now.  Thank you in advance,
and if you need further info, please ask.. 

Chris Platt

platt@coos.dartmouth.edu
"It's not real until you blow some fuses.." -- me.