[net-next] tcp: expose both send and receive intervals for rate sample
diff mbox series

Message ID 1531158819-7901-1-git-send-email-deeptir@mit.edu
State Accepted, archived
Delegated to: David Miller
Headers show
Series
  • [net-next] tcp: expose both send and receive intervals for rate sample
Related show

Commit Message

Deepti Raghavan July 9, 2018, 5:53 p.m. UTC
Congestion control algorithms, which access the rate sample
through the tcp_cong_control function, only have access to the maximum
of the send and receive interval, for cases where the acknowledgment
rate may be inaccurate due to ACK compression or decimation. Algorithms
may want to use send rates and receive rates as separate signals.

Signed-off-by: Deepti Raghavan <deeptir@mit.edu>
---
 include/net/tcp.h   | 2 ++
 net/ipv4/tcp_rate.c | 4 ++++
 2 files changed, 6 insertions(+)

Comments

Neal Cardwell July 9, 2018, 6:09 p.m. UTC | #1
On Mon, Jul 9, 2018 at 1:58 PM Deepti Raghavan <deeptir@mit.edu> wrote:
>
> Congestion control algorithms, which access the rate sample
> through the tcp_cong_control function, only have access to the maximum
> of the send and receive interval, for cases where the acknowledgment
> rate may be inaccurate due to ACK compression or decimation. Algorithms
> may want to use send rates and receive rates as separate signals.
>
> Signed-off-by: Deepti Raghavan <deeptir@mit.edu>
> ---
>  include/net/tcp.h   | 2 ++
>  net/ipv4/tcp_rate.c | 4 ++++
>  2 files changed, 6 insertions(+)

Thanks for re-sending. It does seem to be showing up in patchwork now:
  https://patchwork.ozlabs.org/patch/941532/
And I can confirm I'm able to apply it to net-next.

Acked-by: Neal Cardwell <ncardwell@google.com>

thanks,
neal
Eric Dumazet July 9, 2018, 6:11 p.m. UTC | #2
On 07/09/2018 10:53 AM, Deepti Raghavan wrote:
> Congestion control algorithms, which access the rate sample
> through the tcp_cong_control function, only have access to the maximum
> of the send and receive interval, for cases where the acknowledgment
> rate may be inaccurate due to ACK compression or decimation. Algorithms
> may want to use send rates and receive rates as separate signals.
> 
> Signed-off-by: Deepti Raghavan <deeptir@mit.edu>

Signed-off-by: Eric Dumazet <edumazet@google.com>

(Assuming another CC is coming soon, using this...)

Thanks
David Miller July 12, 2018, 6:02 a.m. UTC | #3
From: Deepti Raghavan <deeptir@mit.edu>
Date: Mon,  9 Jul 2018 17:53:39 +0000

> Congestion control algorithms, which access the rate sample
> through the tcp_cong_control function, only have access to the maximum
> of the send and receive interval, for cases where the acknowledgment
> rate may be inaccurate due to ACK compression or decimation. Algorithms
> may want to use send rates and receive rates as separate signals.
> 
> Signed-off-by: Deepti Raghavan <deeptir@mit.edu>

Applied.

Patch
diff mbox series

diff --git a/include/net/tcp.h b/include/net/tcp.h
index cce3769..f6cb20e 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -954,6 +954,8 @@  struct rate_sample {
 	u32  prior_delivered;	/* tp->delivered at "prior_mstamp" */
 	s32  delivered;		/* number of packets delivered over interval */
 	long interval_us;	/* time for tp->delivered to incr "delivered" */
+	u32 snd_interval_us;	/* snd interval for delivered packets */
+	u32 rcv_interval_us;	/* rcv interval for delivered packets */
 	long rtt_us;		/* RTT of last (S)ACKed packet (or -1) */
 	int  losses;		/* number of packets marked lost upon ACK */
 	u32  acked_sacked;	/* number of packets newly (S)ACKed upon ACK */
diff --git a/net/ipv4/tcp_rate.c b/net/ipv4/tcp_rate.c
index c61240e..4dff40d 100644
--- a/net/ipv4/tcp_rate.c
+++ b/net/ipv4/tcp_rate.c
@@ -146,6 +146,10 @@  void tcp_rate_gen(struct sock *sk, u32 delivered, u32 lost,
 				    rs->prior_mstamp); /* ack phase */
 	rs->interval_us = max(snd_us, ack_us);
 
+	/* Record both segment send and ack receive intervals */
+	rs->snd_interval_us = snd_us;
+	rs->rcv_interval_us = ack_us;
+
 	/* Normally we expect interval_us >= min-rtt.
 	 * Note that rate may still be over-estimated when a spuriously
 	 * retransmistted skb was first (s)acked because "interval_us"