On Sat, 12 Jul 1997 14:32:48 -0700 (PDT) 
 wrstuden@loki.stanford.edu (Bill Studenmund) wrote:

 > We use the Machine-Independant 8530 SCC driver. We thus share code with
 > the sun and sun3 (I think) ports.

sun3, sun3x, and mvme68k.

 > I was able to ftp between my Mac IIsi and a 586 (w/ FIFO UART's)
 > at 57600 without a single overrun. I gto > 10kB/s transfer
 > rates (compression was on). However I DID bring the machine to its
 > knees. I was running dt, and it would take about 3 seconds for a
 > screen to change when I hit Command-#. Also, the clock lost a lot
 > of time (like minutes in tens of minutes of transfer).

...no doubt because of the extremely lame mac68k clock interrupt level.
IIRC, when you're at splsoft*(), you're blocking the clock, which is
going to affect the time, and process scheduling.  (Right?  Doesn't
the clock interrupt at ipl 1?)

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939