mac80211: fix use-after-free in defragmentation
authorJohannes Berg <johannes.berg@intel.com>
Mon, 3 Nov 2014 12:57:46 +0000 (13:57 +0100)
committerZefan Li <lizefan@huawei.com>
Mon, 2 Feb 2015 09:05:18 +0000 (17:05 +0800)
commit b8fff407a180286aa683d543d878d98d9fc57b13 upstream.

Upon receiving the last fragment, all but the first fragment
are freed, but the multicast check for statistics at the end
of the function refers to the current skb (the last fragment)
causing a use-after-free bug.

Since multicast frames cannot be fragmented and we check for
this early in the function, just modify that check to also
do the accounting to fix the issue.

Reported-by: Yosef Khyal <yosefx.khyal@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Zefan Li <lizefan@huawei.com>
net/mac80211/rx.c

index f5ed863..32929b0 100644 (file)
@@ -1486,11 +1486,14 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
        sc = le16_to_cpu(hdr->seq_ctrl);
        frag = sc & IEEE80211_SCTL_FRAG;
 
-       if (likely((!ieee80211_has_morefrags(fc) && frag == 0) ||
-                  is_multicast_ether_addr(hdr->addr1))) {
-               /* not fragmented */
+       if (likely(!ieee80211_has_morefrags(fc) && frag == 0))
+               goto out;
+
+       if (is_multicast_ether_addr(hdr->addr1)) {
+               rx->local->dot11MulticastReceivedFrameCount++;
                goto out;
        }
+
        I802_DEBUG_INC(rx->local->rx_handlers_fragments);
 
        if (skb_linearize(rx->skb))
@@ -1583,10 +1586,7 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
  out:
        if (rx->sta)
                rx->sta->rx_packets++;
-       if (is_multicast_ether_addr(hdr->addr1))
-               rx->local->dot11MulticastReceivedFrameCount++;
-       else
-               ieee80211_led_rx(rx->local);
+       ieee80211_led_rx(rx->local);
        return RX_CONTINUE;
 }