tcp: switch TCP TS option (RFC 7323) to 1ms clock

TCP Timestamps option is defined in RFC 7323

Traditionally on linux, it has been tied to the internal
'jiffies' variable, because it had been a cheap and good enough
generator.

For TCP flows on the Internet, 1 ms resolution would be much better
than 4ms or 10ms (HZ=250 or HZ=100 respectively)

For TCP flows in the DC, Google has used usec resolution for more
than two years with great success [1]

Receive size autotuning (DRS) is indeed more precise and converges
faster to optimal window size.

This patch converts tp->tcp_mstamp to a plain u64 value storing
a 1 usec TCP clock.

This choice will allow us to upstream the 1 usec TS option as
discussed in IETF 97.

[1] https://www.ietf.org/proceedings/97/slides/slides-97-tcpm-tcp-options-for-low-latency-00.pdf

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 10 insertions, 13 deletions

diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c
index cd72b3d3879e..fe9a493d0208 100644
--- a/net/ipv4/tcp_recovery.c
+++ b/net/ipv4/tcp_recovery.c
@@ -17,12 +17,9 @@ static void tcp_rack_mark_skb_lost(struct sock *sk, struct sk_buff *skb)
 	}
 }
 
-static bool tcp_rack_sent_after(const struct skb_mstamp *t1,
-				const struct skb_mstamp *t2,
-				u32 seq1, u32 seq2)
+static bool tcp_rack_sent_after(u64 t1, u64 t2, u32 seq1, u32 seq2)
 {
-	return skb_mstamp_after(t1, t2) ||
-	       (t1->v64 == t2->v64 && after(seq1, seq2));
+	return t1 > t2 || (t1 == t2 && after(seq1, seq2));
 }
 
 /* RACK loss detection (IETF draft draft-ietf-tcpm-rack-01):
@@ -72,14 +69,14 @@ static void tcp_rack_detect_loss(struct sock *sk, u32 *reo_timeout)
 		    scb->sacked & TCPCB_SACKED_ACKED)
 			continue;
 
-		if (tcp_rack_sent_after(&tp->rack.mstamp, &skb->skb_mstamp,
+		if (tcp_rack_sent_after(tp->rack.mstamp, skb->skb_mstamp,
 					tp->rack.end_seq, scb->end_seq)) {
 			/* Step 3 in draft-cheng-tcpm-rack-00.txt:
 			 * A packet is lost if its elapsed time is beyond
 			 * the recent RTT plus the reordering window.
 			 */
-			u32 elapsed = skb_mstamp_us_delta(&tp->tcp_mstamp,
-							  &skb->skb_mstamp);
+			u32 elapsed = tcp_stamp_us_delta(tp->tcp_mstamp,
+							 skb->skb_mstamp);
 			s32 remaining = tp->rack.rtt_us + reo_wnd - elapsed;
 
 			if (remaining < 0) {
@@ -127,16 +124,16 @@ void tcp_rack_mark_lost(struct sock *sk)
  * draft-cheng-tcpm-rack-00.txt
  */
 void tcp_rack_advance(struct tcp_sock *tp, u8 sacked, u32 end_seq,
-		      const struct skb_mstamp *xmit_time)
+		      u64 xmit_time)
 {
 	u32 rtt_us;
 
-	if (tp->rack.mstamp.v64 &&
-	    !tcp_rack_sent_after(xmit_time, &tp->rack.mstamp,
+	if (tp->rack.mstamp &&
+	    !tcp_rack_sent_after(xmit_time, tp->rack.mstamp,
 				 end_seq, tp->rack.end_seq))
 		return;
 
-	rtt_us = skb_mstamp_us_delta(&tp->tcp_mstamp, xmit_time);
+	rtt_us = tcp_stamp_us_delta(tp->tcp_mstamp, xmit_time);
 	if (sacked & TCPCB_RETRANS) {
 		/* If the sacked packet was retransmitted, it's ambiguous
 		 * whether the retransmission or the original (or the prior
@@ -152,7 +149,7 @@ void tcp_rack_advance(struct tcp_sock *tp, u8 sacked, u32 end_seq,
 			return;
 	}
 	tp->rack.rtt_us = rtt_us;
-	tp->rack.mstamp = *xmit_time;
+	tp->rack.mstamp = xmit_time;
 	tp->rack.end_seq = end_seq;
 	tp->rack.advanced = 1;
 }