On Thu, 2013-04-11 at 12:47 +0200, Arend van Spriel wrote:
> On 04/09/2013 12:06 PM, Johannes Berg wrote:
> > On Mon, 2013-04-08 at 11:09 +0200, Arend van Spriel wrote:
> >> +void cfg80211_crit_proto_stopped(struct wireless_dev *wdev)
> >> +{
> >> + struct cfg80211_registered_device *rdev;
> >> +
> >> + rdev = wiphy_to_dev(wdev->wiphy);
> >> + WARN_ON(!rdev->crit_proto_started);
> >> + rdev->crit_proto_started = false;
> >> +}
> >
> > Oh, so you don't want to tell userspace?
>
> Just an observation will looking for an example of netlink event
> messaging to user-space. I noticed that the cfg80211_ft_event() function
> used fixed GFP_KERNEL value. Should event functions always have a gfp_t
> parameter?
Well I guess in general I would add it (unless there are other reasons
it needs to be able to sleep). OTOH, if all drivers always call it in a
context where they can sleep, it's safe to just use GFP_KERNEL, so for
the ft_event it probably doesn't actually matter.
johannes
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
