> Has anyone ever attempted to measure the amount of time interrupts are held
> off?  If not, I've done that kind of instrumentation before (not in netbsd)
> and I might be able to find enough free time to repeat the effort (assuming
> anyone would do anything with the results).

I would be interested to hear about the results.  It would give a rough
estimate of "real-time IO" capabilities of NetBSD in certain applications.
It might also help locating the com problem (if the problem really is that
the interrupts are disabled for such long periods of time).

  -jm