On Fri, Dec 17, 2010 at 11:40 AM, Luis R. Rodriguez <lrodriguez@xxxxxxxxxxx> wrote: > On Fri, Dec 17, 2010 at 1:50 AM, Johannes Berg > <johannes@xxxxxxxxxxxxxxxx> wrote: >> On Thu, 2010-12-16 at 19:23 -0800, Luis R. Rodriguez wrote: >> >>> Â Â Â /* Lets us get back the wiphy on the callback */ >>> Â Â Â int (*reg_notifier)(struct wiphy *wiphy, >>> - Â Â Â Â Â Â Â Â Â Â Â Â struct regulatory_request *request); >>> + Â Â Â Â Â Â Â Â Â Â Â Â struct regulatory_request *request, >>> + Â Â Â Â Â Â Â Â Â Â Â Â u8 dfs_region); >> >> So does that mean we need to implement a reg_notifier if we want to have >> radar detection? Doesn't an empty reg_notifier already alter behaviour > > An empty reg_notifier() should not do anything. Its either through the > reg_notifier() or we add a new callback specifically for DFS. A new > callback may give more flexibility later if we want to pass up more > data. After some more thought I think its best to keep this within the reg_notifier() instead of adding a new callback, the reason being that we are already passing some regulatory data on it, no point in splitting up the calls and having those synch up on themselves later on the driver. But we are missing the alpha2 which I suspect drivers may need to request_firmware() some DFS data for the specific chipset, or do their own lookup on local static data. We don't currently pass the cfg80211_regdomain but I think it makes sense in this case, then the user can just check the alpha2 from it and in case other users need any other data from the regulatory domain they can have at it. I'm going to make this change and submit as PATCH form next. I haven't seen any more feeback on this so I take it that we're OK with this. Luis -- 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
