> This machine uses the WD33C93, right?  If not, than the following is
> sort of pointless... :-)

Nope.  Sorry...  It's the NCR 53C96.  I have the AMD manual for their
clone.  If someone could give me a number for NCR's microelectronics
literature center, that might help, or send me docs ;-)  I have their
number on some old 5380 docs, but neither the 800, or the 303 number
work anymore.  *sigh*

> So how does the current 5380 driver work?  Does it sit in the interrupt
> handler, hogging cycles do access the FIFO at the expense of everything
> else, or is there some "kernel SCSI process" that does this?  I don't
> seem to recall MacBSD locking up the system on SCSI i/o...

It locks up the system on scsi I/O, basically, but only on a
block-to-block basis--interrupts disabled...  Why couldn't Apple
have done DMA and just written the darn MacOS support for it????

-allen

-- 
Allen Briggs - end killing - allen.briggs@vt.edu ** MacBSD == NetBSD/Mac **
= Over the years you swam the ocean following feelings of your own [...]  =
== It's a shame to have to die to put the shadow on our eyes.  We don't  ==
=== want to care. Under the bridge. Over the phone. Wind on the Water.  ===

------------------------------------------------------------------------------