iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

tcp: retire FACK loss detection

Browse Source 
FACK loss detection has been disabled by default and the
successor RACK subsumed FACK and can handle reordering better.
This patch removes FACK to simplify TCP loss recovery.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com>
Reviewed-by: Priyaranjan Jha <priyarjha@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Yuchung Cheng
2017-11-08 13:01:26 -08:00
committed by David S. Miller
parent e4ec138413
commit 713bafea92
10 changed files with 12 additions and 77 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
net/ipv4/tcp_output.c
View File

@ -1257,7 +1257,7 @@ static void tcp_adjust_pcount(struct sock *sk, const struct sk_buff *skb, int de
if (tp->lost_skb_hint &&
before(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(tp->lost_skb_hint)->seq) &&
(tcp_is_fack(tp) || (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)))
(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED))
tp->lost_cnt_hint -= decr;
tcp_verify_left_out(tp);
@ -2961,9 +2961,6 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs)
* retransmitted data is acknowledged. It tries to continue
* resending the rest of the retransmit queue, until either
* we've sent it all or the congestion window limit is reached.
* If doing SACK, the first ACK which comes back for a timeout
* based retransmit packet might feed us FACK information again.
* If so, we use it to avoid unnecessarily retransmissions.
*/
void tcp_xmit_retransmit_queue(struct sock *sk)
{