Izumi Tsutsui wrote:

>In article <437D2462.4020908@tadpole.com>
>garrett_damore@tadpole.com wrote:
>
>  
>
>>I wonder if this is related to an interrupt problem with PCI on the
>>Au1550 core I'm currently debugging.  It manifests with network cards
>>taking 1 full second to respond to ping.  Increasing the ping frequency
>>seems to solve this problem.  However until it is solved, simple things
>>like a single telnet session have unacceptably poor performance.  (1
>>second latency!)
>>    
>>
>
>I don't think softintr functions take such long time (1 sec),
>but it seems softintr requests are lost.
>(ack after 1 sec is caused by the next ping?)
>
>Does your port use MI mips softintr functions? (though evbmips does)
>Does any inputs from serial port (which also triggers softintr)
>also fix the problem?
>---
>Izumi Tsutsui
>  
>

Okay, I feel like a moron now.  It turns out the latency was being
caused by the other side of the network.  (Don't know why that was, but
frankly at this point I don't care too much.)

Sticking a Linux box on the other side and it all works properly. 
(Geez, and to think I wasted nearly a full day debugging _that_
non-problem.)

-- 
Garrett D'Amore                          http://www.tadpolecomputer.com/
Sr. Staff Engineer          Extending the Power of 64-bit UNIX Computing
Tadpole Computer, Inc.                             Phone: (951) 325-2134