5.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet edumazet@google.com
[ Upstream commit e891b36de161fcd96f12ff83667473e5067b9037 ]
If skb_expand_head() returns NULL, skb has been freed and associated dst/idev could also have been freed.
We need to hold rcu_read_lock() to make sure the dst and associated idev are alive.
Fixes: 5796015fa968 ("ipv6: allocate enough headroom in ip6_finish_output2()") Signed-off-by: Eric Dumazet edumazet@google.com Cc: Vasily Averin vasily.averin@linux.dev Reviewed-by: David Ahern dsahern@kernel.org Link: https://patch.msgid.link/20240820160859.3786976-3-edumazet@google.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org (cherry picked from commit e891b36de161fcd96f12ff83667473e5067b9037) Signed-off-by: Harshvardhan Jha harshvardhan.j.jha@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv6/ip6_output.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a88c3988cbbd..45622cc03e96 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -68,11 +68,15 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *
/* Be paranoid, rather than too clever. */ if (unlikely(hh_len > skb_headroom(skb)) && dev->header_ops) { + /* Make sure idev stays alive */ + rcu_read_lock(); skb = skb_expand_head(skb, hh_len); if (!skb) { IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); + rcu_read_unlock(); return -ENOMEM; } + rcu_read_unlock(); }
hdr = ipv6_hdr(skb);