From: Johannes Berg <[email protected]>
Since the cfg80211 and mac80211 radar detection support is
rather broken right now, disable it in the only driver that
advertises support for it. When all the bugs are resolved,
support for it can be restored.
Cc: [email protected] [3.10+]
Signed-off-by: Johannes Berg <[email protected]>
---
drivers/net/wireless/ath/ath9k/init.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c
index 710192e..b9a1ce3 100644
--- a/drivers/net/wireless/ath/ath9k/init.c
+++ b/drivers/net/wireless/ath/ath9k/init.c
@@ -920,7 +920,10 @@ void ath9k_set_hw_capab(struct ath_softc *sc, struct ieee80211_hw *hw)
BIT(NL80211_IFTYPE_ADHOC) |
BIT(NL80211_IFTYPE_MESH_POINT);
hw->wiphy->iface_combinations = if_comb;
- hw->wiphy->n_iface_combinations = ARRAY_SIZE(if_comb);
+ /* radar detection code is rather broken right now,
+ * so for now disable it entirely
+ */
+ hw->wiphy->n_iface_combinations = 1;
}
hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
--
1.8.5.1
On Tue, 2013-12-17 at 18:09 +0100, Simon Wunderlich wrote:
> as discussed already on IRC, what we consider "broken" is: the lockdep issue
> which Kalle (and maybe others?) reported for hwsim, but that does not affect
> ath9k. Probably a hwsim problem only.
It does affect ath9k, but the issue can't actually happen in practice
due to other considerations that lockdep can't really prove. It's still
ugly and we should fix it.
> For the other issue for which Marek Puzyniak sent a patch ("cfg80211: fix dfs
> channel state after stopping AP"), I don't think this is a bug which should be
> fixed, but intential to keep the channel AVAILABLE after leaving the AP on a
> channel. Let's discuss this in the thread for this patch ...
Fair enough.
Let's drop this.
johannes
Hi,
> From: Johannes Berg <[email protected]>
>
> Since the cfg80211 and mac80211 radar detection support is
> rather broken right now, disable it in the only driver that
> advertises support for it. When all the bugs are resolved,
> support for it can be restored.
>
as discussed already on IRC, what we consider "broken" is: the lockdep issue
which Kalle (and maybe others?) reported for hwsim, but that does not affect
ath9k. Probably a hwsim problem only.
For the other issue for which Marek Puzyniak sent a patch ("cfg80211: fix dfs
channel state after stopping AP"), I don't think this is a bug which should be
fixed, but intential to keep the channel AVAILABLE after leaving the AP on a
channel. Let's discuss this in the thread for this patch ...
I don't know if there are other issues, but since it works for me I would
suggest to drop this patch and keep DFS enabled for ath9k.
Thanks,
Simon