igb on a Nehalem system, buildworld stats

[Mars G Miro]

On Fri, Jan 9, 2009 at 6:22 PM, Mars G Miro <spry at anarchy.in.the.ph> wrote:
> On Fri, Jan 9, 2009 at 6:14 PM, Mars G Miro <spry at anarchy.in.the.ph> wrote:
>> On Fri, Jan 9, 2009 at 5:15 PM, Jack Vogel <jfvogel at gmail.com> wrote:
>>>
>> [snip]
>>>> >>
>>>> >> If you have a back to back connection to another NIC on Port 0, no
>>>> >> switch,
>>>> >> does
>>>> >> it still autoneg to 100?
>>>> >>
>>>>
>>>> Connected back to back w/ another box w/ a GigE NIC, it now does
>>>> 1000baseTX:
>>>>
>>>> igb0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
>>>>        options=19b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4>
>>>>        ether 00:30:48:c5:db:e2
>>>>        inet6 fe80::230:48ff:fec5:dbe2%igb0 prefixlen 64 scopeid 0x1
>>>>        inet 192.168.70.2 netmask 0xffffff00 broadcast 192.168.70.255
>>>>        media: Ethernet autoselect (1000baseTX <full-duplex>)
>>>>        status: active
>>>>
>>>> But still not without problems. I hafta ifconfig down/up it several
>>>> times until I can see the other end. W/c is the same for igb1.
>>>
>>>
>>> OK, so you have some switch issue.  What do you mean "see the other end",
>>> if its back to back and boots up I assume it gets link, if you have the
>>> address
>>> assigned in rc.conf, and you run tcpdump on the partner do you see the arp
>>> when it comes online, and at that point can the other side ping it?
>>>
>>
>> By 'see the other end' , I meant that even if It says 1000baseTX, i
>> still can't ping the other end, well not really, as I can now see it
>> gots bad chksums:
>>
>> 1. 001691 00:30:48:c5:db:e2 > ff:ff:ff:ff:ff:ff, ethertype ARP
>> (0x0806), length 60: arp who-has 192.168.70.2 tell 192.168.70.2
>> 1. 511111 00:30:48:c5:db:e2 > 00:30:48:61:d7:f2, ethertype IPv4
>> (0x0800), length 98: (tos 0x0, ttl 64, id 20346, offset 0, flags
>> [none], proto ICMP (1), length 84) 192.168.70.2 > 192.168.70.1: ICMP
>> echo request, id 14852, seq 0, length 64
>> 000012 00:30:48:61:d7:f2 > 00:30:48:c5:db:e2, ethertype IPv4 (0x0800),
>> length 98: (tos 0x0, ttl 64, id 3034, offset 0, flags [none], proto
>> ICMP (1), length 84, bad cksum 0 (->617b)!) 192.168.70.1 >
>> 192.168.70.2: ICMP echo reply, id 14852, seq 0, length 64
>> 1. 001611 00:30:48:c5:db:e2 > 00:30:48:61:d7:f2, ethertype IPv4
>> (0x0800), length 98: (tos 0x0, ttl 64, id 57773, offset 0, flags
>> [none], proto ICMP (1), length 84) 192.168.70.2 > 192.168.70.1: ICMP
>> echo request, id 14852, seq 1, length 64
>> 000011 00:30:48:61:d7:f2 > 00:30:48:c5:db:e2, ethertype IPv4 (0x0800),
>> length 98: (tos 0x0, ttl 64, id 59591, offset 0, flags [none], proto
>> ICMP (1), length 84, bad cksum 0 (->848d)!) 192.168.70.1 >
>> 192.168.70.2: ICMP echo reply, id 14852, seq 1, length 64
>>
>> and this is back to back w/ another box w/ a GigE NIC (nfe, w/c has
>> been in production for some time)
>>
>>> Oh, and what is the link partner hardware?
>>>
>>
>> The switch? it's a Dlink 48-Port DGS-1248T GigE switch.
>>
>> Thanks.
>>
>
> btw,  I tried 200812-CURRENT and CURRENT as of Jan 7 and the behavior
> is still the same :-(
>
>

Hi again,

We installed Windows 7 BETA on it, and altho the 1st NIC behaves the
same way, e.g. if connected to a switch it stays at 100mbps, but if
connected back to back w/ another GigE it can stay at 1.0Gbps, I do
not get any network problems at all, for both igb NICs.

So, something to do w/ the igb driver then ?

Thanks.

>>> Jack
>>>




-- 
cheers
mars