http://bugzilla.novell.com/show_bug.cgi?id=530264 http://bugzilla.novell.com/show_bug.cgi?id=530264#c18 --- Comment #18 from Stefan Quandt <squan@web.de> 2010-07-30 04:12:06 UTC ---
There must some mess in the rfkill synchronization with the radio of the adapter. Why is this only a problem of some ipw2100 users and not all of them?
And yes there must be a rfkill state confustion as the rfkill command queries the file /sys/class/rfkill/rfkill0/name which contained the wrong state while /sys/bus/pci/drivers/ipw2100/0000\:02\:04.0/rf_kill always reports the correct state. Does this mean that ubuntu bug #482291 has been fixed incompletely? On http://marc.info/?l=linux-wireless&m=125796825803032&w=2 Matthew Garrett says libipw unconditionally calls wiphy_unregister, but it's up to the driver to register it in the first place. ipw2100 fails to do so. When did this fix go into openSUSE?
The rfkill issue depends on the HW and BIOS of your computer Yes, the BIOS would be a candidate, escpecially when doing a suspend/resume cycle. But after doing the 'rfkill 0 unblock' fix the rfkill state is not compromized on subsequent suspend/resume cycles and wireless remains connected and just fine.
after disabling radio (FN F2 on my dell) ipw2100 continues to do periodically futile network scans like this: Maybe the FN F2 just turns the radio off and the rfkill interface is not set thus NM does not know the adapter is down and keep scanning. No, i don't have NetworkManager running. So I'm afraid the driver does periodical scanning on its own.
-- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.