mac80211: fix receiving deauth

Marcel reported a warning, which quite obviously comes
from an oversight in the code handling deauth frames,
and which resulted in multiple follow-up warnings due
to this missing handling. This patch adds the missing
deauth handling (telling cfg80211 about it) and also
removes the follow-up warnings since they could happen
due to races even if nothing is wrong. I've explained
the races in the comments.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Reported-by: Marcel Holtmann <marcel@holtmann.org>
Tested-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 76c03da..f60a831 100644 (file)
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2003,6 +2003,9 @@ static void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
        case RX_MGMT_CFG80211_ASSOC:
                cfg80211_send_rx_assoc(sdata->dev, (u8 *) mgmt, skb->len);
                break;
+       case RX_MGMT_CFG80211_DEAUTH:
+               cfg80211_send_deauth(sdata->dev, (u8 *)mgmt, skb->len, NULL);
+               break;
        default:
                WARN(1, "unexpected: %d", rma);
        }
@@ -2498,8 +2501,13 @@ int ieee80211_mgd_deauth(struct ieee80211_sub_if_data *sdata,
                }
        }
 
-       /* cfg80211 should catch this... */
-       if (WARN_ON(!bssid)) {
+       /*
+        * cfg80211 should catch this ... but it's racy since
+        * we can receive a deauth frame, process it, hand it
+        * to cfg80211 while that's in a locked section already
+        * trying to tell us that the user wants to disconnect.
+        */
+       if (!bssid) {
                mutex_unlock(&ifmgd->mtx);
                return -ENOLINK;
        }
@@ -2524,8 +2532,13 @@ int ieee80211_mgd_disassoc(struct ieee80211_sub_if_data *sdata,
 
        mutex_lock(&ifmgd->mtx);
 
-       /* cfg80211 should catch that */
-       if (WARN_ON(&ifmgd->associated->cbss != req->bss)) {
+       /*
+        * cfg80211 should catch this ... but it's racy since
+        * we can receive a disassoc frame, process it, hand it
+        * to cfg80211 while that's in a locked section already
+        * trying to tell us that the user wants to disconnect.
+        */
+       if (&ifmgd->associated->cbss != req->bss) {
                mutex_unlock(&ifmgd->mtx);
                return -ENOLINK;
        }