tg3 and or pci-e bug

["Christensen Tom" <paveraware@hotmail.com>]

I have 3 supermicro systems based on the x6dal-tb2 motherboard.  It has 
built in broadcom 5721 gig-e pci-e nics.  eth0 on these boxes fails whenever 
a decent amount of data is pushed across them (decent being ~100Mb).  At 
this point I can say when it fails I get these error messages in 
/var/log/messages:
Oct  2 19:08:53 office kernel: NETDEV WATCHDOG: eth0: transmit timed out
Oct  2 19:08:53 office kernel: tg3: eth0: transmit timed out, resetting
Oct  2 19:08:53 office kernel: tg3: tg3_stop_block timed out, ofs=1400 
enable_bit=2
Oct  2 19:08:53 office kernel: tg3: tg3_stop_block timed out, ofs=c00 
enable_bit=2
Oct  2 19:08:53 office kernel: tg3: tg3_stop_block timed out, ofs=4800 
enable_bit=2
Oct  2 19:08:53 office kernel: tg3: eth0: Link is down.

I made a cron job to log ifconfig output to a file every minute.  This shows 
that the NIC resets itself at least every couple minutes when data is being 
passed.  The TX/RX stats in ifconfig reset to 0.  The above message in 
/var/log/messages doesn't happen every time the NIC resets like this.  I 
think that the NIC is resetting because of some bug, and sometimes, the 
reset fails and locks the NIC, creating the above messages.  The above only 
happens once or twice a day, the other nic resets happen as I said every 2-3 
minutes.  Is there any information that would be helpful in debugging this 
problem?  Let me know what to run and I'll do it.  Eth1 never has this 
problem, I have pushed 5GB+ onto the box over eth1 and it doesn't blink.
Tom


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/