sundance.o only two ports working

sundance.o only two ports working

[Holger.Woehle]

Hello,
i have a strange problem with two of my machines:
They are identical P4 systems with Intel Chipset with two epro100 adapters
onboard and a d-link dfe-580TX quad ethernet card.
I am using Kernel 2.4.18 and sundance.o v1.07a 7/9/2002.
The problem is, that i only can use the two first ports of the d-link card.
I can set up all four with no problems according to mii-diag and alta-diag.
Each port recognises link up/down and negotiates to the right speed and
flowcontrol.
But when i send traffic over the ports 4&5, at first nothing happens and of
course i can't see any interrupts in /proc/interrupts eth4 eth5 and after a
while i get
the console message:
eth4: Transmit timed out, status c0, resetting...

Having a look at alta-diag i noticed, that the column
" Interrupt status is..." differs between Index#1 / #2 and Index#3 / #4
Index#1 and #2 tells: Interrupt status is 0000: No interrupts pending.
Index#3 and #4 tells: Interrupt status is 0101: Interrupt summary Link status
changed

and after genrating some traffic that line changes at Index#3 & #4 to:

Interrupt status is 0301: Interrupt summary Link status changed Tx DMA done.

The beehavier is identical on both machines.

For better analysing i can send a file including cat /proc/pic, pci-config,
alta-diag -e and alta-diag -m bevor and after sending some pings.
This file is a some more pages long so i don't know if it is ok to send it to
the mailing list.

with regards
Holger

Re: sundance.o only two ports working

[Matthew Hall]

[-- Attachment #1: Type: text/plain, Size: 2642 bytes --]

On Wed, 2002-08-14 at 18:37, Holger.Woehle@arcor.net wrote:
> Hello,
> i have a strange problem with two of my machines:
> They are identical P4 systems with Intel Chipset with two epro100 adapters
> onboard and a d-link dfe-580TX quad ethernet card.
> I am using Kernel 2.4.18 and sundance.o v1.07a 7/9/2002.

I experienced problems with the sundance on the dfe580tx a while back 
too, and I made a patch for 2.4.18 a while back (it on my site (in the
sig)), donald becker then released v1.08 on his site a while back
(www.scyld.com) which fixed the problem (my patch now includes this). 
Please apply this patch and try again, (or try db's original version).
Fyi, the dfe580tx doesn't support mii-diag afaik. The best error check
was to check the output of ifconfig and see whether the RX bytes
equalled the TX bytes (unless running lo, these shouldn't be the same,
and they were the first time I had a problem).
Hope this helps,
Matt

> The problem is, that i only can use the two first ports of the d-link card.
> I can set up all four with no problems according to mii-diag and alta-diag.
> Each port recognises link up/down and negotiates to the right speed and
> flowcontrol.
> But when i send traffic over the ports 4&5, at first nothing happens and of
> course i can't see any interrupts in /proc/interrupts eth4 eth5 and after a
> while i get
> the console message:
> eth4: Transmit timed out, status c0, resetting...
> 
> Having a look at alta-diag i noticed, that the column
> " Interrupt status is..." differs between Index#1 / #2 and Index#3 / #4
> Index#1 and #2 tells: Interrupt status is 0000: No interrupts pending.
> Index#3 and #4 tells: Interrupt status is 0101: Interrupt summary Link status
> changed
> 
> and after genrating some traffic that line changes at Index#3 & #4 to:
> 
> Interrupt status is 0301: Interrupt summary Link status changed Tx DMA done.
> 
> The beehavier is identical on both machines.
> 
> For better analysing i can send a file including cat /proc/pic, pci-config,
> alta-diag -e and alta-diag -m bevor and after sending some pings.
> This file is a some more pages long so i don't know if it is ok to send it to
> the mailing list.
> 
> with regards
> Holger
> 
> 
> 
> -
> 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/
-- 
Matthew Hall -- matt@ecsc.co.uk -- http://people.ecsc.co.uk/~matt/
Sig: Printed on 100% recycled electrons.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 232 bytes --]