Linux wireless drivers development
 help / color / mirror / Atom feed
* [PATCH] mac80211: fix allmulti/promisc behaviour
@ 2007-11-16  1:54 Johannes Berg
  2007-11-16 10:46 ` Johannes Berg
  0 siblings, 1 reply; 2+ messages in thread
From: Johannes Berg @ 2007-11-16  1:54 UTC (permalink / raw)
  To: John W. Linville; +Cc: Michael Wu, linux-wireless

When an interface with promisc/allmulti bit is taken down,
the mac80211 state can become confused. This fixes it by
making mac80211 keep track of all *active* interfaces that
have the promisc/allmulti bit set in the sdata, we sync
the interface bit into sdata at set_multicast_list() time
so this works.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
---
I'm 99.9% sure this is the correct fix but I need to test it.

 net/mac80211/ieee80211.c |   23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

--- everything.orig/net/mac80211/ieee80211.c	2007-11-16 02:43:01.668421822 +0100
+++ everything/net/mac80211/ieee80211.c	2007-11-16 02:49:56.508419055 +0100
@@ -267,6 +267,17 @@ static int ieee80211_open(struct net_dev
 		tasklet_enable(&local->tasklet);
 	}
 
+	/*
+	 * set_multicast_list will be invoked by the networking core
+	 * which will check whether any increments here were done in
+	 * error and sync them down to the hardware as filter flags.
+	 */
+	if (sdata->flags & IEEE80211_SDATA_ALLMULTI)
+		atomic_inc(&local->iff_allmultis);
+
+	if (sdata->flags & IEEE80211_SDATA_PROMISC)
+		atomic_inc(&local->iff_promiscs);
+
 	local->open_count++;
 
 	netif_start_queue(dev);
@@ -284,6 +295,18 @@ static int ieee80211_stop(struct net_dev
 
 	netif_stop_queue(dev);
 
+	/*
+	 * Don't count this interface for promisc/allmulti while it
+	 * is down. dev_mc_unsync() will invoke set_multicast_list
+	 * on the master interface which will sync these down to the
+	 * hardware as filter flags.
+	 */
+	if (sdata->flags & IEEE80211_SDATA_ALLMULTI)
+		atomic_dec(&local->iff_allmultis);
+
+	if (sdata->flags & IEEE80211_SDATA_PROMISC)
+		atomic_dec(&local->iff_promiscs);
+
 	dev_mc_unsync(local->mdev, dev);
 
 	/* down all dependent devices, that is VLANs */



^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] mac80211: fix allmulti/promisc behaviour
  2007-11-16  1:54 [PATCH] mac80211: fix allmulti/promisc behaviour Johannes Berg
@ 2007-11-16 10:46 ` Johannes Berg
  0 siblings, 0 replies; 2+ messages in thread
From: Johannes Berg @ 2007-11-16 10:46 UTC (permalink / raw)
  To: John W. Linville; +Cc: Michael Wu, linux-wireless

[-- Attachment #1: Type: text/plain, Size: 598 bytes --]


On Fri, 2007-11-16 at 02:54 +0100, Johannes Berg wrote:
> When an interface with promisc/allmulti bit is taken down,
> the mac80211 state can become confused. This fixes it by
> making mac80211 keep track of all *active* interfaces that
> have the promisc/allmulti bit set in the sdata, we sync
> the interface bit into sdata at set_multicast_list() time
> so this works.
> 
> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
> ---
> I'm 99.9% sure this is the correct fix but I need to test it.

Done. This patch makes it work correctly, please push for 2.6.24.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2007-11-16 11:56 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-11-16  1:54 [PATCH] mac80211: fix allmulti/promisc behaviour Johannes Berg
2007-11-16 10:46 ` Johannes Berg

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox