After patch:
cfg80211/mac80211: refactor cfg80211_chandef_dfs_required()
start_radar_detection always fail with -EINVAL.
Signed-off-by: Janusz Dziedzic <[email protected]>
---
net/wireless/chan.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/wireless/chan.c b/net/wireless/chan.c
index 84d686e..7a679a6 100644
--- a/net/wireless/chan.c
+++ b/net/wireless/chan.c
@@ -340,6 +340,7 @@ int cfg80211_chandef_dfs_required(struct wiphy *wiphy,
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_P2P_GO:
case NL80211_IFTYPE_MESH_POINT:
+ case NL80211_IFTYPE_UNSPECIFIED:
width = cfg80211_chandef_get_width(chandef);
if (width < 0)
return -EINVAL;
@@ -370,7 +371,6 @@ int cfg80211_chandef_dfs_required(struct wiphy *wiphy,
case NL80211_IFTYPE_AP_VLAN:
case NL80211_IFTYPE_WDS:
case NL80211_IFTYPE_P2P_DEVICE:
- case NL80211_IFTYPE_UNSPECIFIED:
break;
case NUM_NL80211_IFTYPES:
WARN_ON(1);
--
1.7.9.5
On Thu, 2014-05-15 at 12:28 +0300, Eliad Peller wrote:
> On Wed, May 14, 2014 at 5:42 PM, Johannes Berg
> <[email protected]> wrote:
> > On Wed, 2014-05-14 at 13:25 +0200, Janusz Dziedzic wrote:
> >> After patch:
> >> cfg80211/mac80211: refactor cfg80211_chandef_dfs_required()
> >>
> >> start_radar_detection always fail with -EINVAL.
> >
> > Applied, thanks.
> >
> probably a bit late, but i don't really see why passing
> NL80211_IFTYPE_UNSPECIFIED at all, instead of the iftype?
I did briefly wonder about the same, but decided Luca probably knew
better :)
johannes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On Wed, 2014-05-14 at 13:25 +0200, Janusz Dziedzic wrote:
> After patch:
> cfg80211/mac80211: refactor cfg80211_chandef_dfs_required()
>
> start_radar_detection always fail with -EINVAL.
Applied, thanks.
johannes
On Wed, May 14, 2014 at 5:42 PM, Johannes Berg
<[email protected]> wrote:
> On Wed, 2014-05-14 at 13:25 +0200, Janusz Dziedzic wrote:
>> After patch:
>> cfg80211/mac80211: refactor cfg80211_chandef_dfs_required()
>>
>> start_radar_detection always fail with -EINVAL.
>
> Applied, thanks.
>
probably a bit late, but i don't really see why passing
NL80211_IFTYPE_UNSPECIFIED at all, instead of the iftype?
Eliad.
On Thu, 2014-05-15 at 09:51 +0000, Coelho, Luciano wrote:
> On Thu, 2014-05-15 at 11:39 +0200, Johannes Berg wrote:
> > On Thu, 2014-05-15 at 12:28 +0300, Eliad Peller wrote:
> > > On Wed, May 14, 2014 at 5:42 PM, Johannes Berg
> > > <[email protected]> wrote:
> > > > On Wed, 2014-05-14 at 13:25 +0200, Janusz Dziedzic wrote:
> > > >> After patch:
> > > >> cfg80211/mac80211: refactor cfg80211_chandef_dfs_required()
> > > >>
> > > >> start_radar_detection always fail with -EINVAL.
> > > >
> > > > Applied, thanks.
> > > >
> > > probably a bit late, but i don't really see why passing
> > > NL80211_IFTYPE_UNSPECIFIED at all, instead of the iftype?
> >
> > I did briefly wonder about the same, but decided Luca probably knew
> > better :)
>
> Actually I also thought the same thing, but had the impression that
> there were cases when we wouldn't have the iftype when calling this.
> But now I checked both cases where we use NL80211_IFTYPE_UNSPECIFIED and
> it seems that we really don't need it.
>
> I'll send a patch to fix this and revert Janusz's change (unless
> Johannes wants to rebase the mac80211-next tree).
Instead of just reverting Janusz's patch, I'll make
cfg80211_chandef_dfs_required() WARN if UNSPECIFIED is passed.
--
Luca.