IIRC, linux has the possibility to map the physical devices to logical devices in the bootloader, by passing options to the kernel, linked to the base memoryaddress
This may have something to do with it? Kind regards, Nils On Tue, Mar 18, 2008 at 9:49 AM, T. Marvin <[EMAIL PROTECTED]> wrote: > I have a pair of Soekris Net4501's, one purchased at the end of last > year (2007), the next one purchased within the last couple weeks (March > 2008). The newest one had v1.33 Bios (I down graded to both v1.28 and > v1.32 in my testing, more on the reason following), the older one I > presume has v1.32 based on the purchase date, but I can't check because > it's in use. > > I have a fairly stock installation of Slackware 12 installed on them. > It's installed on a CF card. (I'm willing to share my installed image, > I built it for cloning and tweaking. I just need a place to share it > to, it's about 100M bzip'ed, and is an image to drop on a 1G CF card.) > > The error I'm having (on the new one) is that when it boots up, the > ethernet devices will show up as eth0 thru eth2 in the boot logs or > whenever I force removal and reloading (i.e. rmmod, modprobe) of the > 'natsemi' driver, but show up as eth3 thru eth5 when the driver is in > the system. > > Quote from 'dmesg' (the boot log): > natsemi eth0: NatSemi DP8381[56] at 0xa0000000 (0000:00:12.0), > 00:00:24:c9:f2:50, IRQ 10, port TP. > natsemi eth1: NatSemi DP8381[56] at 0xa0001000 (0000:00:13.0), > 00:00:24:c9:f2:51, IRQ 11, port TP. > natsemi eth2: NatSemi DP8381[56] at 0xa0002000 (0000:00:14.0), > 00:00:24:c9:f2:52, IRQ 5, port TP. > > If I try to 'ifconfig' the devices, I get the error of the devices not > being there: > [EMAIL PROTECTED]:~# ifconfig eth0 192.168.1.1 > SIOCSIFADDR: No such device > eth0: ERROR while getting interface flags: No such device > > After much trial and error, I discovered that the ethernet devices were > showing up at eth3 thru eth5. (I configured my install for them showing > up there.) > > [EMAIL PROTECTED]:~# ifconfig (trimmed for brevity) > eth3 Link encap:Ethernet HWaddr 00:00:24:C9:F2:51 > inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0 > Interrupt:11 Base address:0x4000 > > eth4 Link encap:Ethernet HWaddr 00:00:24:C9:F2:50 > inet addr:192.168.5.1 Bcast:192.168.5.255 Mask:255.255.255.0 > Interrupt:10 Base address:0x2000 > > eth5 Link encap:Ethernet HWaddr 00:00:24:C9:F2:52 > inet addr:169.254.66.69 Bcast:169.254.255.255 > Mask:255.255.0.0 > Interrupt:5 Base address:0x6000 > > Note that the base addresses also don't match where the devices were > originally found. > > Now... to make this even weirder: As I was beating my head against the > wall with this, I took a copy off the drive image running on the first > box I built that is working just fine as doesn't have this 'eth#' > remapping anomaly. The new box has the problem with that known good > image. > > I downgraded the Bios on the new box, postulating that it could possibly > be that. Using v1.28, v1.32, and v1.33, they all had the same problem. > So I have two, what I think to be identical, Net4501 boxes, booted them > both from the same drive image (in one test), and one has this oddity. > > Any insight into this would be appreciated. > > ...Trevor -- Simple guidelines to happiness: Work like you don't need the money, Love like your heart has never been broken and Dance like no one can see you.
_______________________________________________ Soekris-tech mailing list [email protected] http://lists.soekris.com/mailman/listinfo/soekris-tech
