(Simulated) console hang?

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

I've picked up my KA630 - or more properly MicroVAX-II - emulator
again.  (I know there are plenty of perfectly good VAX emulators out
there already; I'm doing this more because it's fun for me than because
I think it will be a Great Contribution To The World or the like.)

I have it to the point of booting diskless to single-user.  There isn't
much hardware support yet; autoconfig looks like

MicroVAX II
total memory = 16372 KB
avail memory = 12976 KB
using 230 buffers containing 920 KB of memory
mainbus0 (root)
ibus0 at mainbus0
uba0 at ibus0: Q22
qe0 at uba0 csr 174440 vec 774 ipl 17
qe0: deqna, hardware address 02:00:58:51:41:30
booted from type 96 unit 0 controller 0 adapter 0
boot device: qe0
root on qe0

But I've got a peculiar problem: every once in a while, console
interaction will lock up.  When this happens, the machine is sitting in
its idle loop.  I've added the ability to force console receive and/or
transmit interrupts (ie, cause an `extra' interrupt); that doesn't
help.  But it happens only in the kernel; I never saw it happen until I
got the kernel up far enough to do interrupt-driven console I/O.

Does anyone have any thoughts on what might be doing this, or where
might be worth looking to track it down?  It's a long shot, I know, but
any ideas would be better than what I have now.

If anyone would like to look at (and/or play with) the code, it's in a
git repo: git://git.rodents-montreal.org/Mouse/vax-emul/full is the
thing to clone.  I can also make a checked-out version available by
anonymous FTP, if that would help.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B