On Mon, Mar 21, 2011 at 11:08 AM, Peter Stuge <[email protected]> wrote:
> Luis R. Rodriguez wrote:
>> So ultimately with Atheros cards you should not have to muck with
>> anything,
>
> Except fixing invalid information on cards.

As I noted, regulatory agencies have never address this, or simply
assume this never happens. To be able to make this permissible
regulatory agencies need to make this explicitly allow for this, today
they only explicitly deny this and they make a point to clarify it is
indeed disallowed, at least in the FCC.

> But this of course goes
> exactly the same for all cards! Nothing special with Atheros. (Except
> that maybe noone else than Atheros is using the kernel regdomain
> infrastructure? I haven't looked.)

Every cfg80211 and mac80211 802.11 Linux driver (therefore any new
driver) uses the regulatory framework of the Linux kernel. It was the
point in putting so much effort into the framework. Its also why we
have a public database for it, to let others contribute. I also
encourage the code for regulatory on Linux to be even considered /
used on other OSes, even proprietary OSes. Its permissive licensed
exactly for this purpose.

>> due to the restrictions of the regulatory domain the card was sold for.
>
> The utility I have in mind would help fix the disconnect when a card
> is being used more or less permanently in a reg domain other than the
> one it was sold for, which seems to be fairly common.

But it would then break regulatory if the card changes the CTL index
it needs to use and the card was no calibrated / tested for the new
CTL index for the card. So good luck with that.

  Luis
_______________________________________________
ath9k-devel mailing list
[email protected]
https://lists.ath9k.org/mailman/listinfo/ath9k-devel

Reply via email to