stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Neal Cardwell <ncardwell@google.com>,
	Yuchung Cheng <ycheng@google.com>,
	Soheil Hassas Yeganeh <soheil@google.com>,
	Priyaranjan Jha <priyarjha@google.com>,
	Eric Dumazet <edumazet@google.com>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.9 07/32] tcp_bbr: fix bw probing to raise in-flight data for very small BDPs
Date: Sat,  4 Aug 2018 11:00:57 +0200	[thread overview]
Message-ID: <20180804082650.284138337@linuxfoundation.org> (raw)
In-Reply-To: <20180804082649.908295462@linuxfoundation.org>

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

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

From: Neal Cardwell <ncardwell@google.com>

[ Upstream commit 383d470936c05554219094a4d364d964cb324827 ]

For some very small BDPs (with just a few packets) there was a
quantization effect where the target number of packets in flight
during the super-unity-gain (1.25x) phase of gain cycling was
implicitly truncated to a number of packets no larger than the normal
unity-gain (1.0x) phase of gain cycling. This meant that in multi-flow
scenarios some flows could get stuck with a lower bandwidth, because
they did not push enough packets inflight to discover that there was
more bandwidth available. This was really only an issue in multi-flow
LAN scenarios, where RTTs and BDPs are low enough for this to be an
issue.

This fix ensures that gain cycling can raise inflight for small BDPs
by ensuring that in PROBE_BW mode target inflight values with a
super-unity gain are always greater than inflight values with a gain
<= 1. Importantly, this applies whether the inflight value is
calculated for use as a cwnd value, or as a target inflight value for
the end of the super-unity phase in bbr_is_next_cycle_phase() (both
need to be bigger to ensure we can probe with more packets in flight
reliably).

This is a candidate fix for stable releases.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Priyaranjan Jha <priyarjha@google.com>
Reviewed-by: Eric Dumazet <edumazet@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 |    4 ++++
 1 file changed, 4 insertions(+)

--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -324,6 +324,10 @@ static u32 bbr_target_cwnd(struct sock *
 	/* Reduce delayed ACKs by rounding up cwnd to the next even number. */
 	cwnd = (cwnd + 1) & ~1U;
 
+	/* Ensure gain cycling gets inflight above BDP even for small BDPs. */
+	if (bbr->mode == BBR_PROBE_BW && gain > BBR_UNIT)
+		cwnd += 2;
+
 	return cwnd;
 }
 

  parent reply	other threads:[~2018-08-04 11:05 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-04  9:00 [PATCH 4.9 00/32] 4.9.118-stable review Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 01/32] ipv4: remove BUG_ON() from fib_compute_spec_dst Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 02/32] net: ena: Fix use of uninitialized DMA address bits field Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 03/32] net: fix amd-xgbe flow-control issue Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 04/32] net: lan78xx: fix rx handling before first packet is send Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 05/32] net: mdio-mux: bcm-iproc: fix wrong getter and setter pair Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 06/32] NET: stmmac: align DMA stuff to largest cache line length Greg Kroah-Hartman
2018-08-04  9:00 ` Greg Kroah-Hartman [this message]
2018-08-04  9:00 ` [PATCH 4.9 08/32] xen-netfront: wait xenbus state change when load module manually Greg Kroah-Hartman
2018-08-04  9:00 ` [PATCH 4.9 09/32] netlink: Do not subscribe to non-existent groups Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 10/32] netlink: Dont shift with UB on nlk->ngroups Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 11/32] tcp: do not force quickack when receiving out-of-order packets Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 12/32] tcp: add max_quickacks param to tcp_incr_quickack and tcp_enter_quickack_mode Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 13/32] tcp: do not aggressively quick ack after ECN events Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 14/32] tcp: refactor tcp_ecn_check_ce to remove sk type cast Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 15/32] tcp: add one more quick ack after after ECN events Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 16/32] pinctrl: intel: Read back TX buffer state Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 17/32] sched/wait: Remove the lockless swait_active() check in swake_up*() Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 18/32] bonding: avoid lockdep confusion in bond_get_stats() Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 19/32] inet: frag: enforce memory limits earlier Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 20/32] ipv4: frags: handle possible skb truesize change Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 21/32] net: dsa: Do not suspend/resume closed slave_dev Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 22/32] netlink: Fix spectre v1 gadget in netlink_create() Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 23/32] net: stmmac: Fix WoL for PCI-based setups Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 24/32] squashfs: more metadata hardening Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 26/32] can: ems_usb: Fix memory leak on ems_usb_disconnect() Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 27/32] net: socket: fix potential spectre v1 gadget in socketcall Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 28/32] virtio_balloon: fix another race between migration and ballooning Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 29/32] kvm: x86: vmx: fix vpid leak Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 30/32] crypto: padlock-aes - Fix Nano workaround data corruption Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 31/32] drm/vc4: Reset ->{x, y}_scaling[1] when dealing with uniplanar formats Greg Kroah-Hartman
2018-08-04  9:01 ` [PATCH 4.9 32/32] scsi: sg: fix minor memory leak in error path Greg Kroah-Hartman
2018-08-04  9:30 ` [PATCH 4.9 00/32] 4.9.118-stable review Nathan Chancellor
2018-08-04 12:44   ` Greg Kroah-Hartman
2018-08-04 14:47 ` Guenter Roeck
2018-08-05 11:53 ` Naresh Kamboju

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=20180804082650.284138337@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ncardwell@google.com \
    --cc=priyarjha@google.com \
    --cc=soheil@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).