This isnt the bug however. I have reports from other network card
cases, and that backing rtl8139.c out doesnt change it. It seems to
have appeared about the time the 2nd set of TCP changes went in
Bingo, here is what I think the fix is. Everyone please test this...
Alan, once people have tested and verified this I will send you the
final version of the diff so that the CVS Id's are up to snuff too...
--- net/ipv4/tcp_ipv4.c.~1~ Mon Aug 9 12:05:14 1999
+++ net/ipv4/tcp_ipv4.c Thu Aug 12 08:30:34 1999
@@ -1778,13 +1778,12 @@
case TCP_TW_ACK:
tcp_v4_send_ack(skb, ((struct tcp_tw_bucket *)sk)->snd_nxt,
((struct tcp_tw_bucket *)sk)->rcv_nxt);
- break;
+ goto discard_it;
case TCP_TW_RST:
goto no_tcp_socket;
default:
goto discard_it;
}
- return 0;
}
static void __tcp_v4_rehash(struct sock *sk)
--- net/ipv6/tcp_ipv6.c.~1~ Mon Aug 9 12:05:14 1999
+++ net/ipv6/tcp_ipv6.c Thu Aug 12 08:30:51 1999
@@ -1465,14 +1465,12 @@
case TCP_TW_ACK:
tcp_v6_send_ack(skb, ((struct tcp_tw_bucket *)sk)->snd_nxt,
((struct tcp_tw_bucket *)sk)->rcv_nxt);
- break;
+ goto discard_it;
case TCP_TW_RST:
goto no_tcp_socket;
default:
goto discard_it;
}
-
- return 0;
}
static int tcp_v6_rebuild_header(struct sock *sk)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/