[patch 6/7] mac80211: allow only one IBSS interface

[patch 6/7] mac80211: allow only one IBSS interface

[Joerg Pommnitz]

You write:
> Consider the case of multiple virtual IBSS interfaces
> trying to join the same IBSS.

Well, consider the case that you want to route between
two *different* IBSS networks with a single NIC. This=20
could be really useful and those a nice feature to have.
-- =20
Regards
=20
       Joerg
=20






      Heute schon einen Blick in die Zukunft von E-Mails wagen? Versuch=
en Sie=B4s mit dem neuen Yahoo! Mail. www.yahoo.de/mail
-
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [patch 6/7] mac80211: allow only one IBSS interface

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 711 bytes --]

[Could you please use reply-to-all and not break threads? thanks]

On Thu, 2007-09-27 at 01:08 -0700, Joerg Pommnitz wrote:

> Well, consider the case that you want to route between
> two *different* IBSS networks with a single NIC. This 
> could be really useful and those a nice feature to have.

Yes, but consider the case where it all breaks because you associate to
the same IBSS. IMHO correctness is more important than "nice features".
It would be possible to fix the issue, by keeping track of the beaconing
etc. in software and having the two IBSS interfaces see each other via
software since they can't in hardware, but I'm not interested in
expending that effort right now.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 190 bytes --]

Re: [patch 6/7] mac80211: allow only one IBSS interface

[Joerg Pommnitz]

Well, the Unix and those the Linux way is "Give 'em enough rope...".
Sure, things might break if you don't know what you do, but the
same is true when you do "rm -rf /" as root. I really don't see the
point of protecting root from itself.
=20
-- =20
Regards
=20
       Joerg
=20


----- Urspr=FCngliche Mail ----
Von: Johannes Berg <johannes@sipsolutions.net>
An: Joerg Pommnitz <pommnitz@yahoo.com>
CC: linux-wireless@vger.kernel.org
Gesendet: Donnerstag, den 27. September 2007, 10:50:38 Uhr
Betreff: Re: [patch 6/7] mac80211: allow only one IBSS interface

[Could you please use reply-to-all and not break threads? thanks]

On Thu, 2007-09-27 at 01:08 -0700, Joerg Pommnitz wrote:

> Well, consider the case that you want to route between
> two *different* IBSS networks with a single NIC. This=20
> could be really useful and those a nice feature to have.

Yes, but consider the case where it all breaks because you associate to
the same IBSS. IMHO correctness is more important than "nice features".
It would be possible to fix the issue, by keeping track of the beaconin=
g
etc. in software and having the two IBSS interfaces see each other via
software since they can't in hardware, but I'm not interested in
expending that effort right now.

johannes






      Machen Sie Yahoo! zu Ihrer Startseite. Los geht's:=20
http://de.yahoo.com/set
-
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[patch 0/7] more mac80211 cleanups and fixes

[Johannes Berg]

Most notable is the radiotap fix. Without it, b43 WARN_ON()s when you
don't set the transmit rate.

johannes

[patch 6/7] mac80211: allow only one IBSS interface

[Johannes Berg]

Consider the case of multiple virtual IBSS interfaces
trying to join the same IBSS. They wouldn't see beacons
that the other virtual interface sent of course so this
leads to problems. Disallow it.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>

---
 net/mac80211/ieee80211.c |   21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

--- wireless-dev.orig/net/mac80211/ieee80211.c	2007-09-26 17:21:42.667337342 +0200
+++ wireless-dev/net/mac80211/ieee80211.c	2007-09-26 17:21:51.317337342 +0200
@@ -170,8 +170,25 @@ static int ieee80211_open(struct net_dev
 	list_for_each_entry(nsdata, &local->interfaces, list) {
 		struct net_device *ndev = nsdata->dev;
 
-		if (ndev != dev && ndev != local->mdev && netif_running(ndev) &&
-		    compare_ether_addr(dev->dev_addr, ndev->dev_addr) == 0) {
+		if (ndev != dev && ndev != local->mdev &&
+		    netif_running(ndev)) {
+			/*
+			 * We really only support one IBSS interface, think
+			 * what would happen if they tried to be in the same
+			 * BSS.
+			 */
+			if (sdata->type == IEEE80211_IF_TYPE_IBSS &&
+			    nsdata->type == IEEE80211_IF_TYPE_IBSS &&
+			    netif_running(nsdata->dev))
+				return -EBUSY;
+
+			/*
+			 * The remaining checks are only done for devices
+			 * with the same MAC address.
+			 */
+			if (compare_ether_addr(dev->dev_addr, ndev->dev_addr))
+				continue;
+
 			/*
 			 * check whether it may have the same address
 			 */

--