Steven Stallion wrote:
On Thu, 17 Sep 2009 03:58:22 +0900 (JST), Masa Murayama
I repeated 100m byte file transfer by ftp, but it passed.
Next, I tried netperf and all tests didn't passed because test00 caused
connection timeout in IPv6. After that, all tests caused "no answer from
sever". test00 is based on ftp. solaris wasn't hang or paniced.
Was solaris hanged or paniced in your tests?
I'll try to pass nicdrv.
At this point, I doubt re will pass NICDRV - I need to add in the
additional tx bufs first.
I think I may see a difference in how we are testing. I am not able to
reproduce the hang when used between two hosts on an isolated switch or
crossover. I see the hang when I run the nic against an auto-sensing 10/100
switch (this introduces more variety in the broadcast/multicast traffic
received). Since re does not support autoneg, I have the link configured
for half-duplex; perhaps this could be a source of some issues?
Yes. At 100 half all bets are off.. where NICDRV is concerned.
- Garrett
Steve
_______________________________________________
driver-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/driver-discuss
_______________________________________________
driver-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/driver-discuss
