> On Sat, Feb 22, 2014 at 11:54 AM, Peer, Ilan <ilan.peer@xxxxxxxxx> wrote:
> >> > static bool cfg80211_go_permissive_chan(struct
> >> cfg80211_registered_device *rdev,
> >> > struct ieee80211_channel
> >> > *chan) @@ -621,8 +623,14 @@ static bool
> >> cfg80211_go_permissive_chan(struct cfg80211_registered_device *rdev,
> >> > ASSERT_RTNL();
> >> >
> >> > if (!config_enabled(CONFIG_CFG80211_REG_RELAX_NO_IR) ||
> >> > - !(wiphy->regulatory_flags &
> REGULATORY_ENABLE_RELAX_NO_IR)
> >> ||
> >> > - !(chan->flags & IEEE80211_CHAN_GO_CONCURRENT))
> >> > + !(wiphy->regulatory_flags &
> REGULATORY_ENABLE_RELAX_NO_IR))
> >> > + return false;
> >> > +
> >> > + if (regulatory_indoor_allowed() &&
> >> > + (chan->flags & IEEE80211_CHAN_INDOOR_ONLY))
> >> > + return true;
> >> > +
> >> > + if (!(chan->flags & IEEE80211_CHAN_GO_CONCURRENT))
> >> > return false;
> >>
> >> Shouldn't this check go before the indoor one?
> >>
> >
> > I do not think so, as this relaxations is orthogonal to the GO concurrent one.
> It means that If a channel is enabled but only under indoor and indoor
> operation is hinted the relaxation removes the NO_IR restriction if such
> exists (without the relaxation indoor operation would be allowed but not
> for GO).
>
> Ah yes, the first check remains as this then though:
>
> if (!config_enabled(CONFIG_CFG80211_REG_RELAX_NO_IR) ||
> !(wiphy->regulatory_flags & REGULATORY_ENABLE_RELAX_NO_IR))
> return false;
>
> That seems wrong for the general case.
>
Sorry but I do not follow ... what's the bug?
Thanks,
Ilan.
��.n��������+%������w��{.n�����{���zW����ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
