From 113ab386c7d6625cff284fb10952ff69a58c18a4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 14 Oct 2011 04:57:46 +0000 Subject: [PATCH] ip_gre: dont increase dev->needed_headroom on a live device MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit It seems ip_gre is able to change dev->needed_headroom on the fly. Its is not legal unfortunately and triggers a BUG in raw_sendmsg() skb = sock_alloc_send_skb(sk, ... + LL_ALLOCATED_SPACE(rt->dst.dev) < another cpu change dev->needed_headromm (making it bigger) ... skb_reserve(skb, LL_RESERVED_SPACE(rt->dst.dev)); We end with LL_RESERVED_SPACE() being bigger than LL_ALLOCATED_SPACE() -> we crash later because skb head is exhausted. Bug introduced in commit 243aad83 in 2.6.34 (ip_gre: include route header_len in max_headroom calculation) Reported-by: Elmar Vonlanthen Signed-off-by: Eric Dumazet CC: Timo Teräs CC: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/ip_gre.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index d7bb94c48345..d55110e93120 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -835,8 +835,6 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev if (skb_headroom(skb) < max_headroom || skb_shared(skb)|| (skb_cloned(skb) && !skb_clone_writable(skb, 0))) { struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom); - if (max_headroom > dev->needed_headroom) - dev->needed_headroom = max_headroom; if (!new_skb) { ip_rt_put(rt); dev->stats.tx_dropped++; -- 2.20.1