On Tue, 26 May 1998 21:19:10 -0400 
 Matt Thomas <matt@3am-software.com> wrote:

 > >Please try it with the lastest version of pmax/bus_dma.c (1.2).
 > 
 > After fixing the latest version, ping times are getting reasonable.

Yah, sorry about that :-)

 > matt@baker> mips.ttcp -t -s -f m -n 8192 -u 10.16.176.70
 > ttcp-t: buflen=8192, nbuf=8192, align=16384/0, port=5001  udp  -> 10.16.176.70
 > ttcp-t: socket
 > ttcp-t: 67108864 bytes in 12.60 real seconds = 40.63 Mbit/sec +++
 > ttcp-t: 8198 I/O calls, msec/call = 1.57, calls/sec = 650.62
 > ttcp-t: 0.2user 12.3sys 0:12real 99% 0i+0d 0maxrss 0+2pf 0+144csw

Lot of system time, but that could be lots of things.  What does "ping"
latency look like?

 > Slow compared to my alphas but sure the Ethernet its on.  Tomorrow
 > I get to see if I can boot and run diskless over fddi.

How does this device do under Ultrix or OSF/1 (if you have any DECstations
running OSF/1, that is :-)

In any case, please don't try this on an R4000 until I get a chance to beat
on the other parts :-)

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: +1 650 428 6939