On Wed, Oct 20, 2004 at 03:55:36PM -0700, Phil Oester wrote: > > Ok, crashed again, here is output: > > TCP: empty write queue packets_out = 1 Thanks for the testing. Could you please apply this patch and see if we can get some more information. Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
===== net/ipv4/tcp_input.c 1.81 vs edited ===== --- 1.81/net/ipv4/tcp_input.c 2004-10-04 07:31:39 +10:00 +++ edited/net/ipv4/tcp_input.c 2004-10-21 23:00:54 +10:00 @@ -2492,6 +2492,8 @@ tcp_dec_pcount_approx(&tp->fackets_out, skb); tcp_packets_out_dec(tp, skb); __skb_unlink(skb, skb->list); + BUG_TRAP(!tcp_get_pcount(&tp->packets_out) || + !skb_queue_empty(&sk->sk_write_queue)); sk_stream_free_skb(sk, skb); } ===== net/ipv4/tcp_output.c 1.68 vs edited ===== --- 1.68/net/ipv4/tcp_output.c 2004-10-20 15:13:52 +10:00 +++ edited/net/ipv4/tcp_output.c 2004-10-21 23:01:18 +10:00 @@ -961,6 +961,8 @@ */ tcp_dec_pcount_approx(&tp->fackets_out, next_skb); tcp_packets_out_dec(tp, next_skb); + BUG_TRAP(!tcp_get_pcount(&tp->packets_out) || + !skb_queue_empty(&sk->sk_write_queue)); sk_stream_free_skb(sk, next_skb); } }
