Thanks Greg and Leon

The PHYs was indeed the problem. One down, one one to go:-)

On Friday 13 April 2007 02:32:37 Greg Ungerer wrote:
>
> I have seen this type of behavior (working on 2.4 and not on 2.6) if
> the PHY detection code is not quite right in the AccessLibrary code.
> Check that first, probably ixp400_xscale_sw/src/ethMii/IxEthMii.c or
> in the actual eth driver ixp400_eth.c.

I tested using the same config as the ADI_COYOTE in  ixp400_eth.c, while it 
has 5 PHYs it found mine at address 4 and 5. While my original config used 0 
and 1. Changing that to 4 and 5 works.


On Friday 13 April 2007 12:23:18 Leon Woestenberg wrote:
>The 2.4 kernel is robust against this, the PHY detection in 2.6 is not.

Since it works using 0 and 1 in 2.4, but actually detects as 4 and 5 when  
scanned under 2.6. I'd say that's true.


On Friday 13 April 2007 02:32:37 Greg Ungerer wrote:
>
> Though this wouldn't explain the eepro100 not working.
>
Indeed, this one still remains. SInce it (miss)behaves the same, could it be a 
PHY detection problem on this one too? 


Regards
Knut 

> >>
> >> My board has one i82557/i82558 10/100 Ethernet port(using eepro100
> >> driver) as eth0. And one of the 425s npe port is connected to a regular
> >> ethernet port, while the other is connected to a 4 port switch.


_______________________________________________
uClinux-dev mailing list
[email protected]
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by [email protected]
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

Reply via email to