Thread (2 messages) 2 messages, 2 authors, 2011-01-03

Re: [patch] mac80211: potential null dereference in mesh forwarding

From: Eric Dumazet <hidden>
Date: 2011-01-03 07:45:49
Also in: kernel-janitors, linux-wireless

Le lundi 03 janvier 2011 à 08:43 +0300, Dan Carpenter a écrit :
quoted hunk ↗ jump to hunk
The printk() is supposed to be ratelimited but we should always goto out
when fwd_skb is NULL.  Otherwise it gets dereferenced on the next line.

Signed-off-by: Dan Carpenter <redacted>
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 5e9d3bc..dc8b566 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1831,8 +1831,9 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx)
 
 			fwd_skb = skb_copy(skb, GFP_ATOMIC);
 
-			if (!fwd_skb && net_ratelimit()) {
-				printk(KERN_DEBUG "%s: failed to clone mesh frame\n",
+			if (!fwd_skb) {
+				if (net_ratelimit())
+					printk(KERN_DEBUG "%s: failed to clone mesh frame\n",
 						   sdata->name);
 				goto out;
 			}
Already discovered/coped by Milton Miller.


Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help