Re: [PATCH/RFC] b43: remove input device usage for rfkill

[Larry Finger <Larry.Finger@lwfinger.net>]

Henrique de Moraes Holschuh wrote:
> On Wed, 02 Jul 2008, Johannes Berg wrote:
> I apologise.  Re-reading it, it does look like a rant, indeed.  It wasn't
> supposed to be, but the tone came up nasty and quite defensive.  I didn't
> intend it to.
> 
> The sheer ammount of explanations I have had to type over the whole rfkill
> business is, apparently, getting to me a lot more than expected, and I guess
> that put me in the defensive.  There is no excuse for that.
> 
> So, please accept my appologies on the tone of the last email.  Please
> reconsider reading that email in full, and please try to ignore the rantness
> smell in it.  I did try to explain the technical reasons why some changes
> would be needed in b43.
> 
> I am sorry I didn't send you any patches.  I don't have any access to a b43
> device, and as a rule, I try to avoid modifying code I will not be able to
> test, especially when I don't know the code to begin with and the changes
> are not going to be small, obvious ones.
> 
>> To address one point though: userspace does want a notification when the
>> hw-rfkill is activated so it (e.g. network manager) can ask people to
>> un-rfkill their device.
> 
> I tried to add such notifications directly to the rfkill class.  rfkill now
> issues uevents and calls a notification chain for any state changes.  It has
> been modified to differentiate hw-rfkill from sw-rfkill, and export that
> information to userspace for network-manager and other applications to use.

I am more than happy to test any b43 patches that you generate. My b43 
device has a wireless switch.

Larry