Messages in this thread | | | Subject | Re: [PATCH] tcp: don't use timestamp from repaired skb-s to calculate RTT | From | Eric Dumazet <> | Date | Tue, 12 Aug 2014 05:15:01 -0700 |
| |
On Tue, 2014-08-12 at 13:45 +0400, Andrey Vagin wrote: > We don't know right timestamp for repaired skb-s. Wrong RTT estimations > isn't good, because some congestion modules heavily depends on it. > > This patch adds the TCPCB_REPAIRED flag, which is included in > TCPCB_RETRANS.
...
> + > + /* All packets are restored as if they have > + * already been sent. skb_mstamp isn't set to > + * avoid wrong rtt estimation. > + */ > + if (tp->repair) { > + TCP_SKB_CB(skb)->sacked |= TCPCB_REPAIRED; > + TCP_SKB_CB(skb)->when = tcp_time_stamp; > + } > } > > /* Try to append data to the end of skb. */
Are you sure TCP_SKB_CB(skb)->when needs to be set ?
It should not anymore.
If yes, I believe a comment would help a lot here.
Also, please include this tag to ease stable backports (3.15+) :
Fixes: 431a91242d8d ("tcp: timestamp SYN+DATA messages") Fixes: 740b0f1841f6 ("tcp: switch rtt estimations to usec resolution")
Thanks !
| |