4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <edumazet@xxxxxxxxxx>
[ Upstream commit 04c03114be82194d4a4858d41dba8e286ad1787c ]
soukjin bae reported a crash in tcp_v4_err() handling
ICMP_DEST_UNREACH after tcp_write_queue_head(sk)
returned a NULL pointer.
Current logic should have prevented this :
if (seq != tp->snd_una || !icsk->icsk_retransmits ||
!icsk->icsk_backoff || fastopen)
break;
Problem is the write queue might have been purged
and icsk_backoff has not been cleared.
Signed-off-by: Eric Dumazet <edumazet@xxxxxxxxxx>
Reported-by: soukjin bae <soukjin.bae@xxxxxxxxxxx>
Acked-by: Neal Cardwell <ncardwell@xxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
include/net/tcp.h | 1 +
net/ipv4/tcp.c | 1 -
2 files changed, 1 insertion(+), 1 deletion(-)
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1457,6 +1457,7 @@ static inline void tcp_write_queue_purge
sk_wmem_free_skb(sk, skb);
sk_mem_reclaim(sk);
tcp_clear_all_retrans_hints(tcp_sk(sk));
+ inet_csk(sk)->icsk_backoff = 0;
}
static inline struct sk_buff *tcp_write_queue_head(const struct sock *sk)
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2253,7 +2253,6 @@ int tcp_disconnect(struct sock *sk, int
tp->write_seq += tp->max_window + 2;
if (tp->write_seq == 0)
tp->write_seq = 1;
- icsk->icsk_backoff = 0;
tp->snd_cwnd = 2;
icsk->icsk_probes_out = 0;
tp->packets_out = 0;
