Commit db8da6bb authored by Yuchung Cheng's avatar Yuchung Cheng Committed by David S. Miller
Browse files

tcp: new helper function for RACK loss detection 



Create a new helper tcp_rack_mark_skb_lost to prepare the
upcoming RACK reordering timer support.

Signed-off-by: default avatarYuchung Cheng <ycheng@google.com>
Signed-off-by: default avatarNeal Cardwell <ncardwell@google.com>
Acked-by: default avatarEric Dumazet <edumazet@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 7410191a

net/ipv4/tcp_recovery.c

+14 −7
Original line number Diff line number Diff line
@@ -3,6 +3,19 @@ 


int sysctl_tcp_recovery __read_mostly = TCP_RACK_LOST_RETRANS;



static void tcp_rack_mark_skb_lost(struct sock *sk, struct sk_buff *skb)

{

	struct tcp_sock *tp = tcp_sk(sk);



	tcp_skb_mark_lost_uncond_verify(tp, skb);

	if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) {

		/* Account for retransmits that are lost again */

		TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;

		tp->retrans_out -= tcp_skb_pcount(skb);

		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT);

	}

}



/* Marks a packet lost, if some packet sent later has been (s)acked.

 * The underlying idea is similar to the traditional dupthresh and FACK

 * but they look at different metrics:
@@ -61,13 +74,7 @@ int tcp_rack_mark_lost(struct sock *sk) 
				continue;



			/* skb is lost if packet sent later is sacked */

			tcp_skb_mark_lost_uncond_verify(tp, skb);

			if (scb->sacked & TCPCB_SACKED_RETRANS) {

				scb->sacked &= ~TCPCB_SACKED_RETRANS;

				tp->retrans_out -= tcp_skb_pcount(skb);

				NET_INC_STATS(sock_net(sk),

					      LINUX_MIB_TCPLOSTRETRANSMIT);

			}

			tcp_rack_mark_skb_lost(sk, skb);

		} else if (!(scb->sacked & TCPCB_RETRANS)) {

			/* Original data are sent sequentially so stop early

			 * b/c the rest are all sent after rack_sent

CRA Git | Maintained and supported by SUSTech CRA and CCSE