USB Kernel problem with latest SHR unstable

[Bram Mertens]

On Fri, Apr 24, 2009 at 4:08 PM, Thomas Otterbein <th.otterbein at gmx.net> wrote:
> On Friday 24 April 2009 15:16:38 Nelson Castillo wrote:
>> On Fri, Apr 24, 2009 at 5:42 AM, Thomas Zimmermann
>>
>> <zimmermann at vdm-design.de> wrote:
>> > Am Freitag 24 April 2009 12:20:00 schrieb Robin Paulson:
>> >> 2009/4/24 Ben Thompson <bt4 at york.ac.uk>:
>> >> ok, i'm confused now. why do i want to run this on my host? the
>> >> problem's with the phone, isn't it?
>> >
>> > I don't know why, but the interface name on my host changed. For me it's
>> > now eth1.
>>
>> Check:
>>
>> http://lists.openmoko.org/pipermail/devel/2009-April/005384.html
>>
>> _______________________________________________
>> Openmoko community mailing list
>> community at lists.openmoko.org
>> http://lists.openmoko.org/mailman/listinfo/community
>
> On my Ubuntu 8.10 the shr-unstable behaves exactly as described in the link.
> The X depends on your number of network cards. I use a notebook which has a
> cable and a Wifi ethernet adapter named eth0 and eth1. However the Neo becomes
> eth3 as it connects. I collected the information from the Wiki about USB
> Networking in a script, which i use until I managed to integrate the Neo
> better into Ubuntu and it's various network configuration mechanisms. The
> script has a variable for the interface as the first line. Adapt it to your
> needs and run it after you have connected your Neo to the USB port.
> I use "ifconfig" to check out the known network cards and their configuration.
>

On my Fedora 10 laptop nothing appears to happen anymore when I plug
in the Neo.  No messages in /var/log/messages, nothing in dmesg.

[m8ram at fedora10 ~]$ sudo /sbin/ip link show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP qlen 1000
    link/ether 00:1c:23:01:c4:a0 brd ff:ff:ff:ff:ff:ff
3: wmaster0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc
pfifo_fast state UNKNOWN qlen 1000
    link/ieee802.11 00:1b:77:60:df:64 brd ff:ff:ff:ff:ff:ff
4: wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP qlen 1000
    link/ether 00:1b:77:60:df:64 brd ff:ff:ff:ff:ff:ff
5: pan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN
    link/ether 52:6d:38:6a:32:49 brd ff:ff:ff:ff:ff:ff

I am booting from U-Boot (in NAND) though, the post referenced above
mentions SHR-testing (the distro on the SD I'm booting from) behaving
differently when booted from QI.

I get the same result on my Xubuntu laptop so I fear that something
else got broken with the opkg update, opkg upgrade I performed.

So far I haven't found a description of what is needed on the side of
the FreeRunner itself.

When I check ifconfig -a on the FreeRunner I see an eth0 interface
with no IP address, eth0:avah alias with IP 169.254.7.134 and the
loopback device.
ip link show shows the loopback device and the eth0 interface.

When I check the /etc/network/interfaces it contains a comment right
before the auto usb0 section saying:
# Ethernet/RNDIS gadget (g_ether)
# ... or on host side, usbnet and random hwaddr

ifup usb0 retruns:
ifup: interface usb0 already configured
but ifconfig usb0 returns:
ifconfig usb0: error fetching interface information: Device not found

modprobe g_ether on the FreeRunner fails to locate the module so maybe
some modules are missing?  Or is this module no longer needed?

loading the usbnet module on either the Fedora or the Xubuntu laptop
does not help.

One final note: the hardware is fine, when I start FDOM from the
device's main memory it is still recognized.

Thanks in advance.

Bram