On Mon, Aug 29, 2011 at 6:22 AM, Johannes Berg <johannes@xxxxxxxxxxxxxxxx> wrote: > On Thu, 2011-08-25 at 11:33 -0700, Luis R. Rodriguez wrote: >> From: Luis R. Rodriguez <mcgrof@xxxxxxxxxxxxxxxx> >> >> When you try to create an IBSS network with a lot of nodes >> you typically end up with split networks due to the fact that >> some nodes for one reason or another end up not listening to >> beacons from the node others do with a higher TSF. What ends >> up happening is you split the IBSS network and these do not >> converge. To address this you can force the BSSID as well as >> the SSID but if you just want to specify the SSID you can >> use the ibss_merge knob to force an IBSS merge by focing the >> BSSID to be based on the passed SSID. >> >> Cc: Jouni Malinen <jouni@xxxxxxxxxxxxxxxx> >> Cc: Ivan Seskar <seskar@xxxxxxxxxxxxxxxxxx> >> Cc: Johannes Berg <johannes@xxxxxxxxxxxxxxxx> >> Signed-off-by: Luis R. Rodriguez <mcgrof@xxxxxxxxxxxxxxxx> >> --- >> >> Ivan, I think this is what you were looking for but an alternative to this >> problem is to specify the BSSID every time you want to create your own large >> IBSS network, this would mean that its a requirement and also means we >> acknowledge a limitation by IEEE's adhoc definition. Is there any reason >> forcing a BSSID on the experimenter is not an equally good solution? > > Even setting "fixed-bssid" is acknowledging that already, but I really > see no reason for this to live in the kernel. Agreed, I was just hoping Ivan can test that does indeed fix that issue and we can throw this out as one of those "bogus" features MadWifi had for researchers. 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
