Ethernet card stops working after upgrade to FC1

[Jim Cornette]

Colin Paul Adams wrote:
> It may of course be pure coincidence, but I have just upgraded one of
> my systems from RH9 to FC1, and my ethernet card has stopped working.
> 
> Could it be a coincidental hardware problem, or is it a software problem?
> Here is the output from dmesg:
> 

> PCI: Found IRQ 10 for device 00:0a.0

I qam wondering if IRQ 10 was messed around with Kudzu probing this card.

> 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
> See Documentation/networking/vortex.txt
> 00:0a.0: 3Com PCI 3c900 Boomerang 10Mbps Combo at 0xd000. Vers LK1.1.18-ac


I heard that disabling kudzu during initial bootup will allow your 
ethernet card to work properly. I have a similar card, but did not 
reinstall it back into the machine.

The last time that I tried the card, it fouled up a lot of programs with 
the faulty driver. My boomerang was recognized as a cyclone and would 
not work.

Anyway, from what I heard. It is kudzu leaving the card in an incorrect 
state. After the card is probed, it does not function properly. If it 
was not probed by kudzu, reports have it that it works normally.

I thought that it was caused by the incorrect driver being used for the 
card. I don't know if the driver is failed or Kudzu is fouling up the 
state of the card on exit. This card seems to be pretty popular though.

As stated earlier, there are various bugs filed on this card for RHEL, 
Severn Betas and probably Fedora Core.

If you run dhclient from a root terminal, does your card get an address? 
Mine got the address, but worked incorrectly. Mozilla, GNOME and a few 
other progams would fail to work.

Jim

>  00:a0:24:cf:ad:ab, IRQ 10
>   product code 4650 rev 00.0 date 07-18-96
>   Internal config register is 10302d8, transceivers 0xe138.
>   8K word-wide RAM 3:5 Rx:Tx split, autoselect/10baseT interface.
>   Enabling bus-master transmits and whole-frame receives.
> 00:0a.0: scatter/gather enabled. h/w checksums disabled
> divert: allocating divert_blk for eth0
> eth0: Dropping NETIF_F_SG since no checksum feature.
> divert: freeing divert_blk for eth0
> PCI: Found IRQ 10 for device 00:0a.0
> 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
> See Documentation/networking/vortex.txt
> 00:0a.0: 3Com PCI 3c900 Boomerang 10Mbps Combo at 0xd000. Vers LK1.1.18-ac
>  00:a0:24:cf:ad:ab, IRQ 10
>   product code 4650 rev 00.0 date 07-18-96
>   Internal config register is 102001b, transceivers 0xe138.
>   64K word-wide RAM 1:1 Rx:Tx split, autoselect/10baseT interface.
>   Enabling bus-master transmits and whole-frame receives.
> 00:0a.0: scatter/gather enabled. h/w checksums disabled
> divert: allocating divert_blk for eth0
> eth0: Dropping NETIF_F_SG since no checksum feature.
> eth0: Transmit error, Tx status register d0.
>   Flags; bus-master 1, dirty 1(1) current 1(1)
>   Transmit list 00000000 vs. cea6f240.
>   0: @cea6f200  length 8000002a status 8000002a
>   1: @cea6f240  length 00000000 status 00000000
>   2: @cea6f280  length 00000000 status 00000000
>   3: @cea6f2c0  length 00000000 status 00000000
>   4: @cea6f300  length 00000000 status 00000000
>   5: @cea6f340  length 00000000 status 00000000
>   6: @cea6f380  length 00000000 status 00000000
>   7: @cea6f3c0  length 00000000 status 00000000
>   8: @cea6f400  length 00000000 status 00000000
>   9: @cea6f440  length 00000000 status 00000000
>   10: @cea6f480  length 00000000 status 00000000
>   11: @cea6f4c0  length 00000000 status 00000000
>   12: @cea6f500  length 00000000 status 00000000
>   13: @cea6f540  length 00000000 status 00000000
>   14: @cea6f580  length 00000000 status 00000000
>   15: @cea6f5c0  length 00000000 status 00000000



-- 
You have an unusual equipment for success.  Be sure to use it properly.