git: 99c1c6755de4 - stable/13 - Revert "tcp: finish SACK loss recovery on sudden lack of SACK blocks"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Wed, 14 Sep 2022 01:29:07 UTC
The branch stable/13 has been updated by rscheff: URL: https://cgit.FreeBSD.org/src/commit/?id=99c1c6755de49b2814511d1744ec239446c6ac1d commit 99c1c6755de49b2814511d1744ec239446c6ac1d Author: Richard Scheffenegger <rscheff@FreeBSD.org> AuthorDate: 2022-09-13 15:28:07 +0000 Commit: Richard Scheffenegger <rscheff@FreeBSD.org> CommitDate: 2022-09-13 15:28:07 +0000 Revert "tcp: finish SACK loss recovery on sudden lack of SACK blocks" This reverts commit 3d0b768cc0a07196aa6167b76b32abb928fb03be. --- sys/netinet/tcp_input.c | 39 +++++++++++++++++---------------------- 1 file changed, 17 insertions(+), 22 deletions(-) diff --git a/sys/netinet/tcp_input.c b/sys/netinet/tcp_input.c index 557e27721a67..601b4a63716f 100644 --- a/sys/netinet/tcp_input.c +++ b/sys/netinet/tcp_input.c @@ -287,20 +287,6 @@ kmod_tcpstat_add(int statnum, int val) counter_u64_add(VNET(tcpstat)[statnum], val); } -/* - * Make sure that we only start a SACK loss recovery when - * receiving a duplicate ACK with a SACK block, and also - * complete SACK loss recovery in case the other end - * reneges. - */ -static bool inline -tcp_is_sack_recovery(struct tcpcb *tp, struct tcpopt *to) -{ - return ((tp->t_flags & TF_SACK_PERMIT) && - ((to->to_flags & TOF_SACK) || - (!TAILQ_EMPTY(&tp->snd_holes)))); -} - #ifdef TCP_HHOOK /* * Wrapper for the TCP established input helper hook. @@ -2539,7 +2525,9 @@ tcp_do_segment(struct mbuf *m, struct tcphdr *th, struct socket *so, TCPSTAT_INC(tcps_rcvacktoomuch); goto dropafterack; } - if (tcp_is_sack_recovery(tp, &to)) { + if ((tp->t_flags & TF_SACK_PERMIT) && + ((to.to_flags & TOF_SACK) || + !TAILQ_EMPTY(&tp->snd_holes))) sack_changed = tcp_sack_doack(tp, &to, th->th_ack); else /* @@ -2609,7 +2597,8 @@ tcp_do_segment(struct mbuf *m, struct tcphdr *th, struct socket *so, * duplicating packets or a possible DoS attack. */ if (th->th_ack != tp->snd_una || - (tcp_is_sack_recovery(tp, &to) && + ((tp->t_flags & TF_SACK_PERMIT) && + (to.to_flags & TOF_SACK) && !sack_changed)) break; else if (!tcp_timer_active(tp, TT_REXMT)) @@ -2621,7 +2610,8 @@ tcp_do_segment(struct mbuf *m, struct tcphdr *th, struct socket *so, if (V_tcp_do_prr && IN_FASTRECOVERY(tp->t_flags)) { tcp_do_prr_ack(tp, th, &to); - } else if (tcp_is_sack_recovery(tp, &to) && + } else if ((tp->t_flags & TF_SACK_PERMIT) && + (to.to_flags & TOF_SACK) && IN_FASTRECOVERY(tp->t_flags)) { int awnd; @@ -2694,7 +2684,8 @@ enter_recovery: * snd_ssthresh is already updated by * cc_cong_signal. */ - if (tcp_is_sack_recovery(tp, &to)) { + if ((tp->t_flags & TF_SACK_PERMIT) && + (to.to_flags & TOF_SACK)) { tp->sackhint.prr_delivered = tp->sackhint.sacked_bytes; } else { @@ -2706,7 +2697,8 @@ enter_recovery: tp->sackhint.recover_fs = max(1, tp->snd_nxt - tp->snd_una); } - if (tcp_is_sack_recovery(tp, &to)) { + if ((tp->t_flags & TF_SACK_PERMIT) && + (to.to_flags & TOF_SACK)) { TCPSTAT_INC( tcps_sack_recovery_episode); tp->snd_recover = tp->snd_nxt; @@ -2798,7 +2790,8 @@ enter_recovery: * from the left side. Such partial ACKs should not be * counted as dupacks here. */ - if (tcp_is_sack_recovery(tp, &to) && + if ((tp->t_flags & TF_SACK_PERMIT) && + (to.to_flags & TOF_SACK) && sack_changed) { tp->t_dupacks++; /* limit overhead by setting maxseg last */ @@ -3980,7 +3973,8 @@ tcp_do_prr_ack(struct tcpcb *tp, struct tcphdr *th, struct tcpopt *to) * (del_data) and an estimate of how many bytes are in the * network. */ - if (tcp_is_sack_recovery(tp, to) || + if (((tp->t_flags & TF_SACK_PERMIT) && + (to->to_flags & TOF_SACK)) || (IN_CONGRECOVERY(tp->t_flags) && !IN_FASTRECOVERY(tp->t_flags))) { del_data = tp->sackhint.delivered_data; @@ -4024,7 +4018,8 @@ tcp_do_prr_ack(struct tcpcb *tp, struct tcphdr *th, struct tcpopt *to) * accordingly. */ if (IN_FASTRECOVERY(tp->t_flags)) { - if (tcp_is_sack_recovery(tp, to)) { + if ((tp->t_flags & TF_SACK_PERMIT) && + (to->to_flags & TOF_SACK)) { tp->snd_cwnd = tp->snd_nxt - tp->snd_recover + tp->sackhint.sack_bytes_rexmit + (snd_cnt * maxseg);