> >Though I think a better idea would be to help Scott re-write our interrupt
> >handler so we don't have this problem at all. :-)
> 
> No argument.  I'm just not up for it.  I haven't even done any followup to
> try my original idea yet.

Well, here's 2 cents:

Some of the Quadras have two interrupt modes. The first mode is for MacOS
and the second mode was designed for AUX. Luckily, the AUX mode is setup
"right" with the clock interrupt on top. Perhaps if the kernel senses
it's on a Quadra we can have it set and use that interrupt model.

The Quadra HW model seems to have been quite popular but unfortunately, this
scheme won't work on AV's or II's as they don't have the proper hardware.

There doesn't seem to be a "good" way to do this. If we adopt the serial
scheme mentioned you basically have to give up a serial port to get good clock
timing. :(

Polling's been mentioned as a solution before but I that makes me kind of edgy.
I keep having visions of multiple clock interrupts occuring during a long
SCSI service.

Enough rambling for today... :)