| From | Greg Kroah-Hartman <> | Subject | [PATCH 3.18 020/145] tcp: fix overflow in __tcp_retransmit_skb() | Date | Sun, 16 Apr 2017 12:48:33 +0200 |
| |
3.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <edumazet@google.com>
commit ffb4d6c8508657824bcef68a36b2a0f9d8c09d10 upstream.
If a TCP socket gets a large write queue, an overflow can happen in a test in __tcp_retransmit_skb() preventing all retransmits.
The flow then stalls and resets after timeouts.
Tested:
sysctl -w net.core.wmem_max=1000000000 netperf -H dest -- -s 1000000000
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
--- net/ipv4/tcp_output.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
--- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2481,7 +2481,8 @@ int __tcp_retransmit_skb(struct sock *sk * copying overhead: fragmentation, tunneling, mangling etc. */ if (atomic_read(&sk->sk_wmem_alloc) > - min(sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), sk->sk_sndbuf)) + min_t(u32, sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), + sk->sk_sndbuf)) return -EAGAIN; if (skb_still_in_host_queue(sk, skb))
|