On Sun, 9 Aug 1998 01:34:19 +0200 (MET DST) 
 Lennart Augustsson <augustss@cs.chalmers.se> wrote:

 > > FYI, I believe I've fixed the `warbling' effect (by rearranging the
 > > interrupt priorities to avoid the latency issues -- like I did on the
 > > i386 a long time ago).  I need to test this a bit, but I'll probably
 > > commit it later tonight.

 > Great!  I'm not sure why it helps, but as long as it does...

...because interrupts are typically arranged in a hierarchy to give
devices which can lose data a better chance of servicing their interrupt.

	bio <= net <= tty <= audio

...or something like that :-)

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                            Home: +1 408 866 1912
NAS: M/S 258-5                                       Work: +1 650 604 0935
Moffett Field, CA 94035                             Pager: +1 650 940 5942