[eepro100] single eepro100 occupies eth0 to eth7 (driver v1.13)

Michael Renzmann mrenzmann@web.de
Fri, 20 Apr 2001 12:00:54 +0200

Hi all.

I digged thru the mailinglist archive and found some postings on that error, but there seems no solution for it. So I want to add my description of the problem, hoping that someone could fix that bug.

The box I experienced the troubles with uses an Intel D815EEA motherboard, running an Intel P3 on it. This board uses Intels i815-chipset that also includes an eepro100 interface in it. Using the "out-of-the-box"-eepro100.c that came with the kernel-source of kernel 2.2.19 the interface was not recognized. After exchanging it with v1.13 from scyld.com homepage it was recognized, but the driver initialised 8 ethernet interfaces instead of one. There is another ethernet card in the system with rtl8139 but this can't be the reason for the strange behaviour, I assume. Yesterday time was too short to test the whole stuff without the rtl8139.

The chipset-internal eepro100 was recognized 8 times and occupied eth0 to eth7. eth0 was configured by the startup skripts and worked fine. I did not test "the other devices". But the rtl8139 could not be setup as the eepro ocupied too much interfaces...

Using ifconfig to have a look at every interface showed no base address for eth0 and irq 11. Every other interface also had irq11 and X*0x2000 as base address with X being the interface number. Example: eth1 had base address 0x2000, eth2 had 0x4000, eth3 had 0x6000 and so on.

Is there anything that I can do to contribute to any effort of resolving this misbehaviour of the driver? Donald?

Bye, Mike