It would usually happen soon after trying to do some network traffic
like an ftp from sunsite or such.
This was with 2.0.35 and I found a way to make it happen fairly reliably,
with two etherexpress nics.. I would boot without rc'ing the network up.
Then right after doing "ifconfig eth0 up" I would get a hang.
So I figured, ok.. I can live with just one card I guess.
Well.. I was still getting the hanging.
In syslog I would find after power cycle and fsck :
kernel: eth0: Rx frame at 5644 corrupted, status ffbf, cmd 8bff, next 0644,
pbuf 548b, len 6469
Oodles and oodles of that with different numbers.
Alright, did some dejanewsing... found other people with the
same card and same problems. Found a general low opinion of this
card/driver. Read the config help for this driver in the 2.1.115
kernbel source and saw a mention that is was not recommened at all
previously and that there is a new driver.
Ok.. well I am now running 2.1.115 on that machine with
that driver.. and yipeeeeeee its not crashing.
However... It is giving me syslogs full of :
kernel: eth0: EOF or F not set on received buffer (8000)
and the numbers change to 4000 and 0000.
Pages of that stuff.. I can see it coming in as I type this out.
(its on another machine though.. and it comes a lot faster when I ftp
from that machine.)
So.. is it the driver or the card or both ?
Possibly even the cabling ?
Gonna try two cards again :)
Should I mail this to the guy who maintains/is-developing that driver.
Peter
-- pjordan at blackwire.com 436-0829 Black Wire Media "The meaning of a value is determined by how it is used" : Ousterhout
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html