[PATCH 4.12 060/106] tcp_bbr: introduce bbr_bw_to_pacing_rate() helper

From: Greg Kroah-Hartman
Date: Wed Aug 09 2017 - 12:54:37 EST


4.12-stable review patch. If anyone has any objections, please let me know.

------------------

From: Neal Cardwell <ncardwell@xxxxxxxxxx>


[ Upstream commit f19fd62dafaf1ed6cf615dba655b82fa9df59074 ]

Introduce a helper to convert a BBR bandwidth and gain factor to a
pacing rate in bytes per second. This is a pure refactor, but is
needed for two following fixes.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Neal Cardwell <ncardwell@xxxxxxxxxx>
Signed-off-by: Yuchung Cheng <ycheng@xxxxxxxxxx>
Signed-off-by: Soheil Hassas Yeganeh <soheil@xxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
net/ipv4/tcp_bbr.c | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)

--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -212,6 +212,16 @@ static u64 bbr_rate_bytes_per_sec(struct
return rate >> BW_SCALE;
}

+/* Convert a BBR bw and gain factor to a pacing rate in bytes per second. */
+static u32 bbr_bw_to_pacing_rate(struct sock *sk, u32 bw, int gain)
+{
+ u64 rate = bw;
+
+ rate = bbr_rate_bytes_per_sec(sk, rate, gain);
+ rate = min_t(u64, rate, sk->sk_max_pacing_rate);
+ return rate;
+}
+
/* Pace using current bw estimate and a gain factor. In order to help drive the
* network toward lower queues while maintaining high utilization and low
* latency, the average pacing rate aims to be slightly (~1%) lower than the
@@ -221,10 +231,8 @@ static u64 bbr_rate_bytes_per_sec(struct
*/
static void bbr_set_pacing_rate(struct sock *sk, u32 bw, int gain)
{
- u64 rate = bw;
+ u32 rate = bbr_bw_to_pacing_rate(sk, bw, gain);

- rate = bbr_rate_bytes_per_sec(sk, rate, gain);
- rate = min_t(u64, rate, sk->sk_max_pacing_rate);
if (bbr_full_bw_reached(sk) || rate > sk->sk_pacing_rate)
sk->sk_pacing_rate = rate;
}