From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Priyaranjan Jha <priyarjha@google.com>,
Neal Cardwell <ncardwell@google.com>,
Yuchung Cheng <ycheng@google.com>,
"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.19 35/37] tcp_bbr: refactor bbr_target_cwnd() for general inflight provisioning
Date: Fri, 25 Sep 2020 14:49:03 +0200 [thread overview]
Message-ID: <20200925124726.218739916@linuxfoundation.org> (raw)
In-Reply-To: <20200925124720.972208530@linuxfoundation.org>
From: Priyaranjan Jha <priyarjha@google.com>
commit 232aa8ec3ed979d4716891540c03a806ecab0c37 upstream.
Because bbr_target_cwnd() is really a general-purpose BBR helper for
computing some volume of inflight data as a function of the estimated
BDP, refactor it into following helper functions:
- bbr_bdp()
- bbr_quantization_budget()
- bbr_inflight()
Signed-off-by: Priyaranjan Jha <priyarjha@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/ipv4/tcp_bbr.c | 60 ++++++++++++++++++++++++++++++++++-------------------
1 file changed, 39 insertions(+), 21 deletions(-)
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -315,30 +315,19 @@ static void bbr_cwnd_event(struct sock *
}
}
-/* Find target cwnd. Right-size the cwnd based on min RTT and the
- * estimated bottleneck bandwidth:
+/* Calculate bdp based on min RTT and the estimated bottleneck bandwidth:
*
- * cwnd = bw * min_rtt * gain = BDP * gain
+ * bdp = bw * min_rtt * gain
*
* The key factor, gain, controls the amount of queue. While a small gain
* builds a smaller queue, it becomes more vulnerable to noise in RTT
* measurements (e.g., delayed ACKs or other ACK compression effects). This
* noise may cause BBR to under-estimate the rate.
- *
- * To achieve full performance in high-speed paths, we budget enough cwnd to
- * fit full-sized skbs in-flight on both end hosts to fully utilize the path:
- * - one skb in sending host Qdisc,
- * - one skb in sending host TSO/GSO engine
- * - one skb being received by receiver host LRO/GRO/delayed-ACK engine
- * Don't worry, at low rates (bbr_min_tso_rate) this won't bloat cwnd because
- * in such cases tso_segs_goal is 1. The minimum cwnd is 4 packets,
- * which allows 2 outstanding 2-packet sequences, to try to keep pipe
- * full even with ACK-every-other-packet delayed ACKs.
*/
-static u32 bbr_target_cwnd(struct sock *sk, u32 bw, int gain)
+static u32 bbr_bdp(struct sock *sk, u32 bw, int gain)
{
struct bbr *bbr = inet_csk_ca(sk);
- u32 cwnd;
+ u32 bdp;
u64 w;
/* If we've never had a valid RTT sample, cap cwnd at the initial
@@ -353,7 +342,24 @@ static u32 bbr_target_cwnd(struct sock *
w = (u64)bw * bbr->min_rtt_us;
/* Apply a gain to the given value, then remove the BW_SCALE shift. */
- cwnd = (((w * gain) >> BBR_SCALE) + BW_UNIT - 1) / BW_UNIT;
+ bdp = (((w * gain) >> BBR_SCALE) + BW_UNIT - 1) / BW_UNIT;
+
+ return bdp;
+}
+
+/* To achieve full performance in high-speed paths, we budget enough cwnd to
+ * fit full-sized skbs in-flight on both end hosts to fully utilize the path:
+ * - one skb in sending host Qdisc,
+ * - one skb in sending host TSO/GSO engine
+ * - one skb being received by receiver host LRO/GRO/delayed-ACK engine
+ * Don't worry, at low rates (bbr_min_tso_rate) this won't bloat cwnd because
+ * in such cases tso_segs_goal is 1. The minimum cwnd is 4 packets,
+ * which allows 2 outstanding 2-packet sequences, to try to keep pipe
+ * full even with ACK-every-other-packet delayed ACKs.
+ */
+static u32 bbr_quantization_budget(struct sock *sk, u32 cwnd, int gain)
+{
+ struct bbr *bbr = inet_csk_ca(sk);
/* Allow enough full-sized skbs in flight to utilize end systems. */
cwnd += 3 * bbr_tso_segs_goal(sk);
@@ -368,6 +374,17 @@ static u32 bbr_target_cwnd(struct sock *
return cwnd;
}
+/* Find inflight based on min RTT and the estimated bottleneck bandwidth. */
+static u32 bbr_inflight(struct sock *sk, u32 bw, int gain)
+{
+ u32 inflight;
+
+ inflight = bbr_bdp(sk, bw, gain);
+ inflight = bbr_quantization_budget(sk, inflight, gain);
+
+ return inflight;
+}
+
/* An optimization in BBR to reduce losses: On the first round of recovery, we
* follow the packet conservation principle: send P packets per P packets acked.
* After that, we slow-start and send at most 2*P packets per P packets acked.
@@ -429,7 +446,8 @@ static void bbr_set_cwnd(struct sock *sk
goto done;
/* If we're below target cwnd, slow start cwnd toward target cwnd. */
- target_cwnd = bbr_target_cwnd(sk, bw, gain);
+ target_cwnd = bbr_bdp(sk, bw, gain);
+ target_cwnd = bbr_quantization_budget(sk, target_cwnd, gain);
if (bbr_full_bw_reached(sk)) /* only cut cwnd if we filled the pipe */
cwnd = min(cwnd + acked, target_cwnd);
else if (cwnd < target_cwnd || tp->delivered < TCP_INIT_CWND)
@@ -470,14 +488,14 @@ static bool bbr_is_next_cycle_phase(stru
if (bbr->pacing_gain > BBR_UNIT)
return is_full_length &&
(rs->losses || /* perhaps pacing_gain*BDP won't fit */
- inflight >= bbr_target_cwnd(sk, bw, bbr->pacing_gain));
+ inflight >= bbr_inflight(sk, bw, bbr->pacing_gain));
/* A pacing_gain < 1.0 tries to drain extra queue we added if bw
* probing didn't find more bw. If inflight falls to match BDP then we
* estimate queue is drained; persisting would underutilize the pipe.
*/
return is_full_length ||
- inflight <= bbr_target_cwnd(sk, bw, BBR_UNIT);
+ inflight <= bbr_inflight(sk, bw, BBR_UNIT);
}
static void bbr_advance_cycle_phase(struct sock *sk)
@@ -736,11 +754,11 @@ static void bbr_check_drain(struct sock
bbr->pacing_gain = bbr_drain_gain; /* pace slow to drain */
bbr->cwnd_gain = bbr_high_gain; /* maintain cwnd */
tcp_sk(sk)->snd_ssthresh =
- bbr_target_cwnd(sk, bbr_max_bw(sk), BBR_UNIT);
+ bbr_inflight(sk, bbr_max_bw(sk), BBR_UNIT);
} /* fall through to check if in-flight is already small: */
if (bbr->mode == BBR_DRAIN &&
tcp_packets_in_flight(tcp_sk(sk)) <=
- bbr_target_cwnd(sk, bbr_max_bw(sk), BBR_UNIT))
+ bbr_inflight(sk, bbr_max_bw(sk), BBR_UNIT))
bbr_reset_probe_bw_mode(sk); /* we estimate queue is drained */
}
next prev parent reply other threads:[~2020-09-25 12:54 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-25 12:48 [PATCH 4.19 00/37] 4.19.148-rc1 review Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 01/37] af_key: pfkey_dump needs parameter validation Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 02/37] KVM: fix memory leak in kvm_io_bus_unregister_dev() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 03/37] kprobes: fix kill kprobe which has been marked as gone Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 04/37] mm/thp: fix __split_huge_pmd_locked() for migration PMD Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 05/37] cxgb4: Fix offset when clearing filter byte counters Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 06/37] geneve: add transport ports in route lookup for geneve Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 07/37] hdlc_ppp: add range checks in ppp_cp_parse_cr() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 08/37] ip: fix tos reflection in ack and reset packets Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 09/37] ipv6: avoid lockdep issue in fib6_del() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 10/37] net: DCB: Validate DCB_ATTR_DCB_BUFFER argument Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 11/37] net: dsa: rtl8366: Properly clear member config Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 12/37] net: ipv6: fix kconfig dependency warning for IPV6_SEG6_HMAC Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 13/37] net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 14/37] nfp: use correct define to return NONE fec Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 15/37] tipc: Fix memory leak in tipc_group_create_member() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 16/37] tipc: fix shutdown() of connection oriented socket Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 17/37] tipc: use skb_unshare() instead in tipc_buf_append() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 18/37] bnxt_en: return proper error codes in bnxt_show_temp Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 19/37] bnxt_en: Protect bnxt_set_eee() and bnxt_set_pauseparam() with mutex Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 20/37] net: phy: Avoid NPD upon phy_detach() when driver is unbound Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 21/37] net: qrtr: check skb_put_padto() return value Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 22/37] net: add __must_check to skb_put_padto() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 23/37] ipv4: Update exception handling for multipath routes via same device Greg Kroah-Hartman
2020-09-25 16:51 ` Pavel Machek
2020-09-26 15:46 ` Greg Kroah-Hartman
2020-09-29 12:26 ` Pavel Machek
2020-09-25 12:48 ` [PATCH 4.19 24/37] MAINTAINERS: add CLANG/LLVM BUILD SUPPORT info Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 25/37] kbuild: add OBJSIZE variable for the size tool Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 26/37] Documentation/llvm: add documentation on building w/ Clang/LLVM Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 27/37] Documentation/llvm: fix the name of llvm-size Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 28/37] net: wan: wanxl: use allow to pass CROSS_COMPILE_M68k for rebuilding firmware Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 29/37] net: wan: wanxl: use $(M68KCC) instead of $(M68KAS) " Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 30/37] x86/boot: kbuild: allow readelf executable to be specified Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 4.19 31/37] kbuild: remove AS variable Greg Kroah-Hartman
2020-09-25 12:49 ` [PATCH 4.19 32/37] kbuild: replace AS=clang with LLVM_IAS=1 Greg Kroah-Hartman
2020-09-25 12:49 ` [PATCH 4.19 33/37] kbuild: support LLVM=1 to switch the default tools to Clang/LLVM Greg Kroah-Hartman
2020-09-25 12:49 ` [PATCH 4.19 34/37] mm: memcg: fix memcg reclaim soft lockup Greg Kroah-Hartman
2020-09-25 12:49 ` Greg Kroah-Hartman [this message]
2020-09-25 12:49 ` [PATCH 4.19 36/37] tcp_bbr: adapt cwnd based on ack aggregation estimation Greg Kroah-Hartman
2020-09-25 12:49 ` [PATCH 4.19 37/37] serial: 8250: Avoid error message on reprobe Greg Kroah-Hartman
2020-09-25 17:39 ` [PATCH 4.19 00/37] 4.19.148-rc1 review Pavel Machek
2020-09-26 15:50 ` Greg Kroah-Hartman
2020-09-25 20:02 ` Shuah Khan
2020-09-26 12:35 ` Naresh Kamboju
2020-09-26 15:42 ` Guenter Roeck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200925124726.218739916@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=ncardwell@google.com \
--cc=priyarjha@google.com \
--cc=stable@vger.kernel.org \
--cc=ycheng@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).