From cc9a672ee522d4805495b98680f4a3db5d0a0af9 Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Sun, 12 Feb 2012 18:37:09 +0000 Subject: tcp: allow tcp_sacktag_one() to tag ranges not aligned with skbs This commit allows callers of tcp_sacktag_one() to pass in sequence ranges that do not align with skb boundaries, as tcp_shifted_skb() needs to do in an upcoming fix in this patch series. In fact, now tcp_sacktag_one() does not need to depend on an input skb at all, which makes its semantics and dependencies more clear. Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 36 ++++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) (limited to 'net/ipv4/tcp_input.c') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 976034f82320..4e8a81fda65c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1307,25 +1307,26 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb, return in_sack; } -static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk, - struct tcp_sacktag_state *state, +/* Mark the given newly-SACKed range as such, adjusting counters and hints. */ +static u8 tcp_sacktag_one(struct sock *sk, + struct tcp_sacktag_state *state, u8 sacked, + u32 start_seq, u32 end_seq, int dup_sack, int pcount) { struct tcp_sock *tp = tcp_sk(sk); - u8 sacked = TCP_SKB_CB(skb)->sacked; int fack_count = state->fack_count; /* Account D-SACK for retransmitted packet. */ if (dup_sack && (sacked & TCPCB_RETRANS)) { if (tp->undo_marker && tp->undo_retrans && - after(TCP_SKB_CB(skb)->end_seq, tp->undo_marker)) + after(end_seq, tp->undo_marker)) tp->undo_retrans--; if (sacked & TCPCB_SACKED_ACKED) state->reord = min(fack_count, state->reord); } /* Nothing to do; acked frame is about to be dropped (was ACKed). */ - if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) + if (!after(end_seq, tp->snd_una)) return sacked; if (!(sacked & TCPCB_SACKED_ACKED)) { @@ -1344,13 +1345,13 @@ static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk, /* New sack for not retransmitted frame, * which was in hole. It is reordering. */ - if (before(TCP_SKB_CB(skb)->seq, + if (before(start_seq, tcp_highest_sack_seq(tp))) state->reord = min(fack_count, state->reord); /* SACK enhanced F-RTO (RFC4138; Appendix B) */ - if (!after(TCP_SKB_CB(skb)->end_seq, tp->frto_highmark)) + if (!after(end_seq, tp->frto_highmark)) state->flag |= FLAG_ONLY_ORIG_SACKED; } @@ -1368,8 +1369,7 @@ static u8 tcp_sacktag_one(const struct sk_buff *skb, struct sock *sk, /* Lost marker hint past SACKed? Tweak RFC3517 cnt */ if (!tcp_is_fack(tp) && (tp->lost_skb_hint != NULL) && - before(TCP_SKB_CB(skb)->seq, - TCP_SKB_CB(tp->lost_skb_hint)->seq)) + before(start_seq, TCP_SKB_CB(tp->lost_skb_hint)->seq)) tp->lost_cnt_hint += pcount; if (fack_count > tp->fackets_out) @@ -1425,7 +1425,11 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, } /* We discard results */ - tcp_sacktag_one(skb, sk, state, dup_sack, pcount); + tcp_sacktag_one(sk, state, + TCP_SKB_CB(skb)->sacked, + TCP_SKB_CB(skb)->seq, + TCP_SKB_CB(skb)->end_seq, + dup_sack, pcount); /* Difference in this won't matter, both ACKed by the same cumul. ACK */ TCP_SKB_CB(prev)->sacked |= (TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS); @@ -1664,10 +1668,14 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, break; if (in_sack) { - TCP_SKB_CB(skb)->sacked = tcp_sacktag_one(skb, sk, - state, - dup_sack, - tcp_skb_pcount(skb)); + TCP_SKB_CB(skb)->sacked = + tcp_sacktag_one(sk, + state, + TCP_SKB_CB(skb)->sacked, + TCP_SKB_CB(skb)->seq, + TCP_SKB_CB(skb)->end_seq, + dup_sack, + tcp_skb_pcount(skb)); if (!before(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp))) -- cgit v1.2.3-59-g8ed1b From daef52bab1fd26e24e8e9578f8fb33ba1d0cb412 Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Sun, 12 Feb 2012 18:37:10 +0000 Subject: tcp: fix range tcp_shifted_skb() passes to tcp_sacktag_one() Fix the newly-SACKed range to be the range of newly-shifted bytes. Previously - since 832d11c5cd076abc0aa1eaf7be96c81d1a59ce41 - tcp_shifted_skb() incorrectly called tcp_sacktag_one() with the start and end sequence numbers of the skb it passes in set to the range just beyond the range that is newly-SACKed. This commit also removes a special-case adjustment to lost_cnt_hint in tcp_shifted_skb() since the pre-existing adjustment of lost_cnt_hint in tcp_sacktag_one() now properly handles this things now that the correct start sequence number is passed in. Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'net/ipv4/tcp_input.c') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 4e8a81fda65c..8116d06e042c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1388,6 +1388,9 @@ static u8 tcp_sacktag_one(struct sock *sk, return sacked; } +/* Shift newly-SACKed bytes from this skb to the immediately previous + * already-SACKed sk_buff. Mark the newly-SACKed bytes as such. + */ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, struct tcp_sacktag_state *state, unsigned int pcount, int shifted, int mss, @@ -1395,12 +1398,11 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *prev = tcp_write_queue_prev(sk, skb); + u32 start_seq = TCP_SKB_CB(skb)->seq; /* start of newly-SACKed */ + u32 end_seq = start_seq + shifted; /* end of newly-SACKed */ BUG_ON(!pcount); - if (skb == tp->lost_skb_hint) - tp->lost_cnt_hint += pcount; - TCP_SKB_CB(prev)->end_seq += shifted; TCP_SKB_CB(skb)->seq += shifted; @@ -1424,12 +1426,11 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, skb_shinfo(skb)->gso_type = 0; } - /* We discard results */ - tcp_sacktag_one(sk, state, - TCP_SKB_CB(skb)->sacked, - TCP_SKB_CB(skb)->seq, - TCP_SKB_CB(skb)->end_seq, - dup_sack, pcount); + /* Adjust counters and hints for the newly sacked sequence range but + * discard the return value since prev is already marked. + */ + tcp_sacktag_one(sk, state, TCP_SKB_CB(skb)->sacked, + start_seq, end_seq, dup_sack, pcount); /* Difference in this won't matter, both ACKed by the same cumul. ACK */ TCP_SKB_CB(prev)->sacked |= (TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS); -- cgit v1.2.3-59-g8ed1b From 0af2a0d0576205dda778d25c6c344fc6508fc81d Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Mon, 13 Feb 2012 20:22:08 +0000 Subject: tcp: fix tcp_shifted_skb() adjustment of lost_cnt_hint for FACK This commit ensures that lost_cnt_hint is correctly updated in tcp_shifted_skb() for FACK TCP senders. The lost_cnt_hint adjustment in tcp_sacktag_one() only applies to non-FACK senders, so FACK senders need their own adjustment. This applies the spirit of 1e5289e121372a3494402b1b131b41bfe1cf9b7f - except now that the sequence range passed into tcp_sacktag_one() is correct we need only have a special case adjustment for FACK. Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net/ipv4/tcp_input.c') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 8116d06e042c..53c8ce4046b2 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1403,6 +1403,10 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, BUG_ON(!pcount); + /* Adjust hint for FACK. Non-FACK is handled in tcp_sacktag_one(). */ + if (tcp_is_fack(tp) && (skb == tp->lost_skb_hint)) + tp->lost_cnt_hint += pcount; + TCP_SKB_CB(prev)->end_seq += shifted; TCP_SKB_CB(skb)->seq += shifted; -- cgit v1.2.3-59-g8ed1b