netdev - [RFC PATCHv2 net-next 2/2] tcp: TLP restart (TLPR)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <b1ed93c04ebc853c5bd4366392147ed8212fc706.1449561786.git.per.hurtig@kau.se>
Date:	Tue,  8 Dec 2015 10:19:43 +0100
From:	Per Hurtig <per.hurtig@....se>
To:	davem@...emloft.net, edumazet@...gle.com, ncardwell@...gle.com,
	nanditad@...gle.com, tom@...bertland.com, ycheng@...gle.com,
	viro@...iv.linux.org.uk, fw@...len.de, mleitner@...hat.com,
	daniel@...earbox.net, willemb@...gle.com,
	ilpo.jarvinen@...sinki.fi, pasi.sarolahti@....fi,
	stephen@...workplumber.org, netdev@...r.kernel.org
Cc:	anna.brunstrom@....se, apetlund@...ula.no, michawe@....uio.no,
	mohammad.rajiullah@....se, Per Hurtig <per.hurtig@....se>
Subject: [RFC PATCHv2 net-next 2/2] tcp: TLP restart (TLPR)

This patch implements the TLP restart modification (TLPR). When data is
ACKed, and TLP's PTO timer is restarted, the time elapsed since the last
outstanding segment was transmitted is subtracted from the calculated RTO
value to not unnecessarily delay loss probes.

Signed-off-by: Per Hurtig <per.hurtig@....se>
---
 include/net/tcp.h     |  2 +-
 net/ipv4/tcp_input.c  |  2 +-
 net/ipv4/tcp_output.c | 12 ++++++++++--
 3 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 833efb7..14c0888 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -568,7 +568,7 @@ void tcp_push_one(struct sock *, unsigned int mss_now);
 void tcp_send_ack(struct sock *sk);
 void tcp_send_delayed_ack(struct sock *sk);
 void tcp_send_loss_probe(struct sock *sk);
-bool tcp_schedule_loss_probe(struct sock *sk);
+bool tcp_schedule_loss_probe(struct sock *sk, bool restart);
 
 /* tcp_input.c */
 void tcp_resume_early_retransmit(struct sock *sk);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2870af3..952c4cd 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3660,7 +3660,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	}
 
 	if (icsk->icsk_pending == ICSK_TIME_RETRANS)
-		tcp_schedule_loss_probe(sk);
+		tcp_schedule_loss_probe(sk, true);
 	tcp_update_pacing_rate(sk);
 	return 1;
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index a800cee..b89ba50 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2135,7 +2135,7 @@ repair:
 
 		/* Send one loss probe per tail loss episode. */
 		if (push_one != 2)
-			tcp_schedule_loss_probe(sk);
+			tcp_schedule_loss_probe(sk, false);
 		is_cwnd_limited |= (tcp_packets_in_flight(tp) >= tp->snd_cwnd);
 		tcp_cwnd_validate(sk, is_cwnd_limited);
 		return false;
@@ -2143,10 +2143,11 @@ repair:
 	return !tp->packets_out && tcp_send_head(sk);
 }
 
-bool tcp_schedule_loss_probe(struct sock *sk)
+bool tcp_schedule_loss_probe(struct sock *sk, bool restart)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct sk_buff *skb = tcp_write_queue_head(sk);
 	u32 timeout, tlp_time_stamp, rto_time_stamp;
 	u32 rtt = usecs_to_jiffies(tp->srtt_us >> 3);
 
@@ -2186,6 +2187,13 @@ bool tcp_schedule_loss_probe(struct sock *sk)
 	if (tp->packets_out == 1)
 		timeout = max_t(u32, timeout,
 				(rtt + (rtt >> 1) + TCP_DELACK_MAX));
+	if (sysctl_tcp_timer_restart & TCP_TIMER_TLPRESTART && restart && skb) {
+		const u32 rto_time_stamp = tcp_skb_timestamp(skb);
+		s32 delta = (s32)(tcp_time_stamp - rto_time_stamp);
+
+		if (delta > 0 && timeout > delta)
+			timeout -= delta;
+	}
 	timeout = max_t(u32, timeout, msecs_to_jiffies(10));
 
 	/* If RTO is shorter, just schedule TLP in its place. */
-- 
1.9.1


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html