Intel PRO/1000 Gigabit Ethernet Adapters

[Benjie Chen]

hi


robert and i were hte one who worked with the e1000 driver over here at mit...
we have not had this kind of problems...

you could check a few things: are you using our e1000.o file? is the
click version matches the driver (that is are they from the same
cvs tar ball)? in polldevice there is a place where it checks this
but verifying the pollable field of the eth data structure.

did you try with just e1000 card?

let me know.

benjie


On Wed, May 30, 2001 at 10:56:31AM +0200, Brecht Vermeulen wrote:
> 
> Hi,
> 
> I've also tried to use Intel gbit cards with the latest cvs source
> (yesterday with click-1.2.0 and the e1000 driver from cvs there was a
> kernel oops), but now click gives the following errors :
> 
> May 30 09:50:02 helios100 kernel: line 13: While configuring
> `PollDevice at 4 :: PollDevice':
> May 30 09:50:02 helios100 kernel:   device `eth4' not pollable, use
> FromDevice instead
> May 30 09:50:02 helios100 kernel: line 26: While configuring
> `PollDevice at 11 :: PollDevice':
> May 30 09:50:02 helios100 kernel:   device `eth5' not pollable, use
> FromDevice instead
> 
> Could this be caused by conflicting IRQs (there are 6 interfaces in the
> system) ?:
> (eth0-eth3 are a DLINK DFE-570 4 port card, with tulip driver and
> polling)
> eth4-5 are 2 gbit NICs with e1000 driver
> 
> eth0      Link encap:Ethernet  HWaddr 00:80:C8:C9:65:50
>           inet addr:10.10.10.100  Bcast:10.10.10.255  Mask:255.255.255.0
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:3370 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:2479 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:100
>           Interrupt:12 Base address:0x7000
> 
> eth2      Link encap:Ethernet  HWaddr 00:80:C8:C9:65:52
>           inet addr:10.0.1.100  Bcast:10.0.1.255  Mask:255.255.255.0
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:328 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:577 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:100
>           Interrupt:11 Base address:0x7800
>  
> eth3      Link encap:Ethernet  HWaddr 00:80:C8:C9:65:53
>           inet addr:10.0.2.100  Bcast:10.0.2.255  Mask:255.255.255.0
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:45 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:44 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:100
>           Interrupt:5 Base address:0x7c00
>  
> eth4      Link encap:Ethernet  HWaddr 00:03:47:71:45:EB
>           inet addr:10.0.5.100  Bcast:10.0.5.255  Mask:255.255.255.0
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:193813 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:193813 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:100
>           Interrupt:11
>  
> eth5      Link encap:Ethernet  HWaddr 00:03:47:71:58:EF
>           inet addr:10.0.6.100  Bcast:10.0.6.255  Mask:255.255.255.0
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:193805 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:193805 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:100
>           Interrupt:10
> 
> If I use the Fromdevices instead, the machine freezes after a very short
> time.
> 
> If I ifconfig eth2 down, so there are no conflicting IRQs (I don't know
> if this is really enough, the card further generates interrupts ?), then
> the machine freezes immediately when installing the config.
> 
> However the same e1000 module works perfectly under normal Linux.
> 
> Any clue for this or any things I could test ? (we've invested recently
> in 4 gbit cards, approx. 3000$, so it would be very interesting to get
> them working ;-) ).
> 
> best regards,
> Brecht

-- 
benjie chen
benjie at lcs.mit.edu