On Wed, Oct 29, 2008 at 12:59 AM, Johannes Berg <johannes@xxxxxxxxxxxxxxxx> wrote: > On Tue, 2008-10-28 at 23:54 +0200, Tomas Winkler wrote: > >> >> Am I missing something? This seems very strange! What specifies that >> >> when an SSID is set on an interface the user wants to scan for that one? >> >> The interface might just have been associated previously to that SSID, >> >> now disassociated and network manager is asking for a scan to get >> >> information about networks to let the user select a new one. >> > >> > NACK, this is a heuristic that helps connection to configured SSID. It >> > doesn't mean that other ssid won't be scanned. >> > This code might be revised again when multiple SSID scan will be implemented >> > Tomas >> > >> Okay I've only now have seen the real reason for this. The SSID >> removal from the driver API. It's much better argument for removing >> this code anyway have to think how this will be substituted. > > :) > Unless somebody can come up with a good argument for requiring the SSID? > I don't see it being necessary the way beaconing and everything is > designed in mac80211 now, and this leaves only the BSSID in the > interface config which I'll move to BSS config and then remove the > interface config thing completely. > > Also, doesn't ieee80211_sta_config_auth already do exactly this > heuristic thing? If it doesn't know about the BSS it'll initiate a scan > for the SSID it's looking for? Yep, you are right. Tomas > > 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
