tcp: cleanup sk_tx_skb_cache before reuse
authorEric Dumazet <edumazet@google.com>
Fri, 29 Mar 2019 19:46:17 +0000 (12:46 -0700)
committerDavid S. Miller <davem@davemloft.net>
Fri, 29 Mar 2019 20:16:44 +0000 (13:16 -0700)
TCP stack relies on the fact that a freshly allocated skb
has skb->cb[] and skb_shinfo(skb)->tx_flags cleared.

When recycling tx skb, we must ensure these fields are cleared.

Fixes: 472c2e07eef0 ("tcp: add one skb cache for tx")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv4/tcp.c

index 82bd707c03472f2cebb1a90d5f1c13acc821468f..603e770d59b3db96adca9602319d2f6970a56285 100644 (file)
@@ -872,6 +872,8 @@ struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp,
                        sk->sk_tx_skb_cache = NULL;
                        pskb_trim(skb, 0);
                        INIT_LIST_HEAD(&skb->tcp_tsorted_anchor);
+                       skb_shinfo(skb)->tx_flags = 0;
+                       memset(TCP_SKB_CB(skb), 0, sizeof(struct tcp_skb_cb));
                        return skb;
                }
        }