[PATCH net] tcp_bbr: fix pacing_gain to always be unity when using lt_bw

[PATCH net] tcp_bbr: fix pacing_gain to always be unity when using lt_bw

[Neal Cardwell]

This commit fixes the pacing_gain to remain at BBR_UNIT (1.0) when
using lt_bw and returning from the PROBE_RTT state to PROBE_BW.

Previously, when using lt_bw, upon exiting PROBE_RTT and entering
PROBE_BW the bbr_reset_probe_bw_mode() code could sometimes randomly
end up with a cycle_idx of 0 and hence have bbr_advance_cycle_phase()
set a pacing gain above 1.0. In such cases this would result in a
pacing rate that is 1.25x higher than intended, potentially resulting
in a high loss rate for a little while until we stop using the lt_bw a
bit later.

This commit is a stable candidate for kernels back as far as 4.9.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Reported-by: Beyers Cronje <bcronje@gmail.com>
---
 net/ipv4/tcp_bbr.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
index 8322f26e770e..25c5a0b60cfc 100644
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -481,7 +481,8 @@ static void bbr_advance_cycle_phase(struct sock *sk)
 
 	bbr->cycle_idx = (bbr->cycle_idx + 1) & (CYCLE_LEN - 1);
 	bbr->cycle_mstamp = tp->delivered_mstamp;
-	bbr->pacing_gain = bbr_pacing_gain[bbr->cycle_idx];
+	bbr->pacing_gain = bbr->lt_use_bw ? BBR_UNIT :
+					    bbr_pacing_gain[bbr->cycle_idx];
 }
 
 /* Gain cycling: cycle pacing gain to converge to fair share of available bw. */
@@ -490,8 +491,7 @@ static void bbr_update_cycle_phase(struct sock *sk,
 {
 	struct bbr *bbr = inet_csk_ca(sk);
 
-	if ((bbr->mode == BBR_PROBE_BW) && !bbr->lt_use_bw &&
-	    bbr_is_next_cycle_phase(sk, rs))
+	if (bbr->mode == BBR_PROBE_BW && bbr_is_next_cycle_phase(sk, rs))
 		bbr_advance_cycle_phase(sk);
 }
 
-- 
2.16.0.rc1.238.g530d649a79-goog

Re: [PATCH net] tcp_bbr: fix pacing_gain to always be unity when using lt_bw

[David Miller]

From: Neal Cardwell <ncardwell@google.com>
Date: Wed, 31 Jan 2018 15:43:05 -0500

> This commit fixes the pacing_gain to remain at BBR_UNIT (1.0) when
> using lt_bw and returning from the PROBE_RTT state to PROBE_BW.
> 
> Previously, when using lt_bw, upon exiting PROBE_RTT and entering
> PROBE_BW the bbr_reset_probe_bw_mode() code could sometimes randomly
> end up with a cycle_idx of 0 and hence have bbr_advance_cycle_phase()
> set a pacing gain above 1.0. In such cases this would result in a
> pacing rate that is 1.25x higher than intended, potentially resulting
> in a high loss rate for a little while until we stop using the lt_bw a
> bit later.
> 
> This commit is a stable candidate for kernels back as far as 4.9.
> 
> Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
> Signed-off-by: Neal Cardwell <ncardwell@google.com>
> Signed-off-by: Yuchung Cheng <ycheng@google.com>
> Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
> Reported-by: Beyers Cronje <bcronje@gmail.com>

Applied and queued up for -stable, thanks Neal.