I've commited a number of low-level interrupt handling changes that should
resolve the problems that some folks have had with the zs driver.  Thanks
to Bill Studenmund for his analysis and Hauke Fath for testing Bill's
theories, I was able to come up with a solution to the problem in fairly
short order.

This should not adversely affect any part of the system, as interrupt
handling itself is functionally a strict superset of what it has been. 
The zs driver was fixed as a side effect of converting to the new
interrupt code.  (I love happy `coincidences'... :-)

--scott

-> Get your official NetBSD-1.3.2 CDROM set today! http://www.netbsd.com/ <-