Question of CSA callback

Question of CSA callback

[Arend van Spriel]

Hi Johannes,

Recently I have taken a look at csa switch callback. After looking more  
closely I decided to let mac80211 handle this itself, but I noticed  
something strange in the interface so maybe you can clarify this.

void (*channel_switch)(struct ieee80211_hw *hw,
		       struct ieee80211_channel_switch *ch_switch);

void ieee80211_chswitch_done(struct ieee80211_vif *vif, bool success);

Basically we get a channel switch callback for iee80211_hw and for the  
ieee80211_chswitch_done() the driver has to provide a vif. Should the  
channel switch also operate on a vif?

Gr. AvS
-- 
"The most merciful thing in the world, I think, is the inability of the  
human
mind to correlate all its contents." - "The Call of Cthulhu"

Re: Question of CSA callback

[Johannes Berg]

On Thu, 2011-04-07 at 13:42 +0200, Arend van Spriel wrote:
> Hi Johannes,
> 
> Recently I have taken a look at csa switch callback. After looking more  
> closely I decided to let mac80211 handle this itself, but I noticed  
> something strange in the interface so maybe you can clarify this.
> 
> void (*channel_switch)(struct ieee80211_hw *hw,
> 		       struct ieee80211_channel_switch *ch_switch);
> 
> void ieee80211_chswitch_done(struct ieee80211_vif *vif, bool success);
> 
> Basically we get a channel switch callback for iee80211_hw and for the  
> ieee80211_chswitch_done() the driver has to provide a vif. Should the  
> channel switch also operate on a vif?

Ah, well, no, not right now anyway -- mac80211 doesn't support
multi-channel operation yet.

However, I think this kind of hw offload isn't necessary for your device
at all.

johannes

Re: Question of CSA callback

[Arend van Spriel]

On Thu, 07 Apr 2011 13:52:08 +0200, Johannes Berg  
<johannes@sipsolutions.net> wrote:

> On Thu, 2011-04-07 at 13:42 +0200, Arend van Spriel wrote:
>> Hi Johannes,
>>
>> Recently I have taken a look at csa switch callback. After looking more
>> closely I decided to let mac80211 handle this itself, but I noticed
>> something strange in the interface so maybe you can clarify this.
>>
>> void (*channel_switch)(struct ieee80211_hw *hw,
>> 		       struct ieee80211_channel_switch *ch_switch);
>>
>> void ieee80211_chswitch_done(struct ieee80211_vif *vif, bool success);
>>
>> Basically we get a channel switch callback for iee80211_hw and for the
>> ieee80211_chswitch_done() the driver has to provide a vif. Should the
>> channel switch also operate on a vif?
>
> Ah, well, no, not right now anyway -- mac80211 doesn't support
> multi-channel operation yet.
>
> However, I think this kind of hw offload isn't necessary for your device
> at all.

Indeed I concluded we do not need this, but I just wanted to clarify the  
intended use of the interface.

Gr. AvS
-- 
"The most merciful thing in the world, I think, is the inability of the  
human
mind to correlate all its contents." - "The Call of Cthulhu"