ar6k in andy-tracking

Robert Piasek robert.piasek at member.fsf.org
Sun Feb 22 11:31:03 CET 2009


On Sunday 22 February 2009 00:48:33 you wrote:
> insmod ...
> ifconfig -a
> dmesg
>
> That way, it also includes the ifconfig -a, in case the badness
> develops after the insmod.


OK testing using stable c96f61f70656e8e2760a29791254a70fdade59e5

I had to blacklist ar6000 module to avoid loading it by udev.

Module loaded ok, and eth0 interface appeared. Well everything as it should 
be. 

Than I removed blacklist from udev and allowed module to be loaded 
automatically by udev (the way it always use to work). This time ifconfig -a 
also showed eth0, but few seconds later it was gone again. (I couldn't spot it 
earlier since I had X starting up and slowing everything down. For testing I 
disabled X this time).

The only messages I could find in logs were:

[21474712.070000] neo1973-pm-bt neo1973-pm-bt.0: __gta02_pm_bt_toggle_radio 0
[21474712.090000] usb 1-1: USB disconnect, address 2
[21474712.090000] btusb_intr_complete: hci0 urb c7a32720 failed to resubmit 
(19)
[21474712.100000] neo1973-pm-bt neo1973-pm-bt.0: __gta02_pm_bt_toggle_radio 0
[21474712.165000] btusb_send_frame: hci0 urb c7a323a0 submission failed
[21474713.610000] ar6000_wow interrupt
[21474715.635000] mmc1: card 0001 removed
[21474715.840000] s3c2440-sdi s3c2440-sdi: powered down.
[21474715.840000] s3c2440-sdi s3c2440-sdi: powered down.


OK, so maybe something from userspace is doing some nasty little magic here. I 
decided to check boot without FSO frameworkd. Everything was ok, no problems 
here. When I started frameword - my network card is gone.

I found out, that frameword is taking the device down.

Should frameworkd go that far? It can power the device down when not used, but 
should it remove the interface? I'm not sure if that's the default killswitch 
behaviour or not.

Anyway, thanks for your time and hints Werner.

Regards,
Rob


PS.
>By the way, this is inconsistent with a  grep ar6000  not finding
>anything. So I think you may have sent the wrong dmesg - or your
>system is in a lot more trouble :)
Heh this was just me thinking FR with 5 days uptime can show complete dmesg 
since boot... (>.<)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.openmoko.org/pipermail/openmoko-kernel/attachments/20090222/3fc06e96/attachment.pgp 


More information about the openmoko-kernel mailing list