1.6_BETA5 - lpt0/lpa0 problems?

[Neil Hoggarth <njh%kernighan.demon.co.uk@localhost>]

I've just done a fresh installation of NetBSD/cats 1.6_BETA5 (2002-08-09
daily snapshot) to a new disk in my Cats system.

I've got most things configured and running with minimal hassle, but I
seem to have problems setting up printing via the parallel port. I've
copied over the /etc/printcap from my working NetBSD/arm32 1.5.3_ALPHA
disk. This includes ":lp=/dev/lpa0:".

If I submit print jobs then they just sit in the queue. The status string
for the print queue is "waiting for lp to become ready (offline ?)". If I
try and redirect output directly to /dev/lpa0 or /dev/lpt0 then I get a
"/dev/lpa0: Device not configured" message from my shell (bash).

I'd normally associate this with having left out the device in question
from the kernel. I'm using the GENERIC kernel from the binary sets that I
installed. dmesg includes "lpt0 at isa0 port 0x378-0x37b irq 7" (same
output as with the 1.5.3 kernel), and grepping the _CFG_ lines from
/netbsd shows an appropriate looking "lpt0 at isa? port 0x378 irq 7" line.

The only thing that I can think of is that the MAKEDEV script has given me
/dev/lp{a,t}0 files which don't have the correct device numbers, but I
don't know what the correct numbers would be. I currently have:

crw-------  1 root  wheel  8, 128 Aug 12 14:17 /dev/lpa0
crw-------  1 root  wheel  8,   0 Aug 12 14:16 /dev/lpt0

(which matches my working arm32 1.5.3 system).

Anybody got any ideas about what is up?

If I am making any obvious mistakes then pointers would be appreciated.
Alternatively, if there is actually something wrong with the lpt driver or
MAKEDEV script for Cats in 1.6_BETA then I need some help characterizing
the problem so I can file a PR.

Regards,

Neil.