[ltp] Re: thinkpad-acpi (T6x/R6x/X6x): requesting testers for wireless switch behaviour

[Henrique de Moraes Holschuh]

On Fri, 09 May 2008, Rainer Rehak wrote:
> Second, i use the madwifi-ng drivers with the norfkill option for my
> wireless. This disables the hardware switch (got it? =)).

The Atheros WLAN cards (at least the older mini-PCI ones) typically
don't connect to the hardware switch.  If they did, one should not be
able to override it AFAIK.

It would require a very special kind of braindamage to add a register to
the hardware that makes it *ignore* the hardware rfkill signal.  If
Atheros does this, then I am *really* happy I got Intel radios instead.

Now, for driver-based rfkill, which every WLAN driver does (on top of
any possible hardware-based rfkill signal), that would make sense.  But
I hate the madwify{,-ng} code so much, I won't look into it to check
what the heck is happening or that will put me off doing kernel work for
a week.  If someone else wants to do it, please reply with your
findings.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh