On Tue, 29 Jul 1997, Erik E. Fair wrote:

> The Sun3 and Mac68k NetBSD ports already have PDMA,
> but they still suffer from the overflows from time to time and expanding the
> PDMA buffer won't help - we're not getting to the hardware fifo fast enough.

Erik,

I agree with your goals, but I wanted to make a wee clarification:  the
mac68k `zs' driver almost never has silo overflows these days, due to the
SCC interrupt being at a higher IPL than virtually all others.

For reference, my home LAN is gatewayed through a IIx running NetBSD, and
I can't find a single instance in my /var/log/messages* files. :-)

Now, clock interrupts... those are a problem.  I don't know that we can
easily craft a solution for that one, given the hardware constraints.

--scott