Re: [PATCH 1/4] mac80211: OCB mode + join and leave handling

[Johannes Berg <johannes@sipsolutions.net>]

On Thu, 2014-10-16 at 18:33 +0200, Rostislav Lisovy wrote:

> You are right. I hope the following is a reasonable solution (in form of
> a patch to my previous patch; comment stolen from some prehistoric
> version of mesh.c):
> 
> @@ -127,6 +127,9 @@ void ieee80211_ocb_work(struct ieee80211_sub_if_data *sdata)
>         struct ieee80211_if_ocb *ifocb = &sdata->u.ocb;
>         struct sta_info *sta;
>  
> +       if (!netif_running(sdata->dev))
> +               return;

Not sure, it seems you should check "is it operating in OCB mode"? OTOH,
when it's not operating but still around it probably doesn't matter?

> @@ -229,6 +232,13 @@ int ieee80211_ocb_leave(struct ieee80211_sub_if_data *sdata)
>         skb_queue_purge(&sdata->skb_queue);
>  
>         del_timer_sync(&sdata->u.ocb.housekeeping_timer);
> +       /*
> +        * If the timer fired while we waited for it, it will have
> +        * requeued the work. Now the work will be running again
> +        * but will not rearm the timer again because it checks
> +        * whether the interface is running, which, at this point,
> +        * it no longer is.
> +        */

Well, the comment is wrong, since leave() can and will be done while the
interface is running.

> > This isn't safe - ocb_rx_no_sta() used GFP_KERNEL, that's clearly not
> > allowed in this context. But it does answer my previous question about
> > the function being exported - I had assumed that you wouldn't call it
> > here since it would be unsafe :)
> 
> A call to sta_info_alloc(sdata, addr, GFP_ATOMIC);
> in ieee80211_ocb_rx_no_sta() should solve this.

Yeah, I guess so, didn't check in detail now.

johannes