Masa Murayama wrote:
> Steve,
> 
> I could duplicate the issue by netperf. The system hanged easily.
> I think I found the rootcause.

I made the adjustments as you suggested, however I am still seeing the
hang. One thing I did notice last night while debugging, is that somehow
the rep->re_rx_next pointer is being corrupted (probably from a bad
value in the rx header) - this causes the recv logic to veer off course
and begin reading from invalid locations in the NIC memory.

Were you able to fix the hang locally?

Cheers,

Steve
_______________________________________________
driver-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/driver-discuss

Reply via email to