summaryrefslogtreecommitdiffstats
path: root/net
diff options
context:
space:
mode:
authorJohannes Berg <johannes@sipsolutions.net>2007-08-28 23:01:55 +0200
committerDavid S. Miller <davem@sunset.davemloft.net>2007-10-11 01:48:56 +0200
commit8dc06a1c6112fef7616d26e0b001455b5d8c3c5e (patch)
treee7e37614847c2e7b26834a0a97bfddbac38661e0 /net
parent[MAC80211]: filter locally-originated multicast frames (diff)
downloadlinux-8dc06a1c6112fef7616d26e0b001455b5d8c3c5e.tar.xz
linux-8dc06a1c6112fef7616d26e0b001455b5d8c3c5e.zip
[MAC80211]: improve key selection comment
When I changed the code there I forgot to mention what happens with multicast frames in a regular BSS and keep wondering myself if the code is correct. Add appropriate comments. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Michael Wu <flamingice@sourmilk.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net')
-rw-r--r--net/mac80211/rx.c7
1 files changed, 7 insertions, 0 deletions
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 969be3adda7a..75a1e204b9b6 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -327,8 +327,15 @@ ieee80211_rx_h_load_key(struct ieee80211_txrx_data *rx)
* frames can also use key indizes like GTKs. Hence, if we don't
* have a PTK/STK we check the key index for a WEP key.
*
+ * Note that in a regular BSS, multicast frames are sent by the
+ * AP only, associated stations unicast the frame to the AP first
+ * which then multicasts it on their behalf.
+ *
* There is also a slight problem in IBSS mode: GTKs are negotiated
* with each station, that is something we don't currently handle.
+ * The spec seems to expect that one negotiates the same key with
+ * every station but there's no such requirement; VLANs could be
+ * possible.
*/
if (!(rx->fc & IEEE80211_FCTL_PROTECTED))