[PATCH net-next 0/2] mptcp: add receive buffer auto-tuning

[PATCH net-next 0/2] mptcp: add receive buffer auto-tuning

[Florian Westphal]

First patch extends the test script to allow for reproducible results.
Second patch adds receive auto-tuning.  Its based on what TCP is doing,
only difference is that we use the largest RTT of any of the subflows
and that we will update all subflows with the new value.

Else, we get spurious packet drops because the mptcp work queue might
not be able to move packets from subflow socket to master socket
fast enough.  Without the adjustment, TCP may drop the packets because
the subflow socket is over its rcvbuffer limit.

Florian Westphal (2):
      selftests: mptcp: add option to specify size of file to transfer
      mptcp: add receive buffer auto-tuning

 net/mptcp/protocol.c                               | 123 +++++++++++++++++++--
 net/mptcp/protocol.h                               |   7 ++
 net/mptcp/subflow.c                                |   5 +-
 tools/testing/selftests/net/mptcp/mptcp_connect.sh |  52 ++++++---
 4 files changed, 166 insertions(+), 21 deletions(-)

[PATCH net-next 1/2] selftests: mptcp: add option to specify size of file to transfer

[Florian Westphal]

The script generates two random files that are then sent via tcp and
mptcp connections.

In order to compare throughput over consecutive runs add an option
to provide the file size on the command line: "-f 128000".

Also add an option, -t, to enable tcp tests. This is useful to
compare throughput of mptcp connections and tcp connections.

Example: run tests with a 4mb file size, 300ms delay 0.01% loss,
default gso/tso/gro settings and with large write/blocking io:

mptcp_connect.sh -t -f $((4 * 1024 * 1024)) -d 300 -l 0.01%  -r 0 -e "" -m mmap

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 .../selftests/net/mptcp/mptcp_connect.sh      | 52 ++++++++++++++-----
 1 file changed, 39 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh
index acf02e156d20..8f7145c413b9 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh
@@ -3,7 +3,7 @@
 
 time_start=$(date +%s)
 
-optstring="S:R:d:e:l:r:h4cm:"
+optstring="S:R:d:e:l:r:h4cm:f:t"
 ret=0
 sin=""
 sout=""
@@ -21,6 +21,8 @@ testmode=""
 sndbuf=0
 rcvbuf=0
 options_log=true
+do_tcp=0
+filesize=0
 
 if [ $tc_loss -eq 100 ];then
 	tc_loss=1%
@@ -40,9 +42,11 @@ usage() {
 	echo -e "\t-e: ethtool features to disable, e.g.: \"-e tso -e gso\" (default: randomly disable any of tso/gso/gro)"
 	echo -e "\t-4: IPv4 only: disable IPv6 tests (default: test both IPv4 and IPv6)"
 	echo -e "\t-c: capture packets for each test using tcpdump (default: no capture)"
+	echo -e "\t-f: size of file to transfer in bytes (default random)"
 	echo -e "\t-S: set sndbuf value (default: use kernel default)"
 	echo -e "\t-R: set rcvbuf value (default: use kernel default)"
 	echo -e "\t-m: test mode (poll, sendfile; default: poll)"
+	echo -e "\t-t: also run tests with TCP (use twice to non-fallback tcp)"
 }
 
 while getopts "$optstring" option;do
@@ -94,6 +98,12 @@ while getopts "$optstring" option;do
 	"m")
 		testmode="$OPTARG"
 		;;
+	"f")
+		filesize="$OPTARG"
+		;;
+	"t")
+		do_tcp=$((do_tcp+1))
+		;;
 	"?")
 		usage $0
 		exit 1
@@ -449,20 +459,25 @@ make_file()
 {
 	local name=$1
 	local who=$2
+	local SIZE=$filesize
+	local ksize
+	local rem
 
-	local SIZE TSIZE
-	SIZE=$((RANDOM % (1024 * 8)))
-	TSIZE=$((SIZE * 1024))
+	if [ $SIZE -eq 0 ]; then
+		local MAXSIZE=$((1024 * 1024 * 8))
+		local MINSIZE=$((1024 * 256))
 
-	dd if=/dev/urandom of="$name" bs=1024 count=$SIZE 2> /dev/null
+		SIZE=$(((RANDOM * RANDOM + MINSIZE) % MAXSIZE))
+	fi
 
-	SIZE=$((RANDOM % 1024))
-	SIZE=$((SIZE + 128))
-	TSIZE=$((TSIZE + SIZE))
-	dd if=/dev/urandom conv=notrunc of="$name" bs=1 count=$SIZE 2> /dev/null
+	ksize=$((SIZE / 1024))
+	rem=$((SIZE - (ksize * 1024)))
+
+	dd if=/dev/urandom of="$name" bs=1024 count=$ksize 2> /dev/null
+	dd if=/dev/urandom conv=notrunc of="$name" bs=1 count=$rem 2> /dev/null
 	echo -e "\nMPTCP_TEST_FILE_END_MARKER" >> "$name"
 
-	echo "Created $name (size $TSIZE) containing data sent by $who"
+	echo "Created $name (size $(du -b "$name")) containing data sent by $who"
 }
 
 run_tests_lo()
@@ -497,9 +512,11 @@ run_tests_lo()
 		return 1
 	fi
 
-	# don't bother testing fallback tcp except for loopback case.
-	if [ ${listener_ns} != ${connector_ns} ]; then
-		return 0
+	if [ $do_tcp -eq 0 ]; then
+		# don't bother testing fallback tcp except for loopback case.
+		if [ ${listener_ns} != ${connector_ns} ]; then
+			return 0
+		fi
 	fi
 
 	do_transfer ${listener_ns} ${connector_ns} MPTCP TCP ${connect_addr} ${local_addr}
@@ -516,6 +533,15 @@ run_tests_lo()
 		return 1
 	fi
 
+	if [ $do_tcp -gt 1 ] ;then
+		do_transfer ${listener_ns} ${connector_ns} TCP TCP ${connect_addr} ${local_addr}
+		lret=$?
+		if [ $lret -ne 0 ]; then
+			ret=$lret
+			return 1
+		fi
+	fi
+
 	return 0
 }
 
-- 
2.26.2

[PATCH net-next 2/2] mptcp: add receive buffer auto-tuning

[Florian Westphal]

When mptcp is used, userspace doesn't read from the tcp (subflow)
socket but from the parent (mptcp) socket receive queue.

skbs are moved from the subflow socket to the mptcp rx queue either from
'data_ready' callback (if mptcp socket can be locked), a work queue, or
the socket receive function.

This means tcp_rcv_space_adjust() is never called and thus no receive
buffer size auto-tuning is done.

An earlier (not merged) patch added tcp_rcv_space_adjust() calls to the
function that moves skbs from subflow to mptcp socket.
While this enabled autotuning, it also meant tuning was done even if
userspace was reading the mptcp socket very slowly.

This adds mptcp_rcv_space_adjust() and calls it after userspace has
read data from the mptcp socket rx queue.

Its very similar to tcp_rcv_space_adjust, with two differences:

1. The rtt estimate is the largest one observed on a subflow
2. The rcvbuf size and window clamp of all subflows is adjusted
   to the mptcp-level rcvbuf.

Otherwise, we get spurious drops at tcp (subflow) socket level if
the skbs are not moved to the mptcp socket fast enough.

Before:
time mptcp_connect.sh -t -f $((4*1024*1024)) -d 300 -l 0.01% -r 0 -e "" -m mmap
[..]
ns4 MPTCP -> ns3 (10.0.3.2:10108      ) MPTCP   (duration 40823ms) [ OK ]
ns4 MPTCP -> ns3 (10.0.3.2:10109      ) TCP     (duration 23119ms) [ OK ]
ns4 TCP   -> ns3 (10.0.3.2:10110      ) MPTCP   (duration  5421ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10111) MPTCP   (duration 41446ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10112) TCP     (duration 23427ms) [ OK ]
ns4 TCP   -> ns3 (dead:beef:3::2:10113) MPTCP   (duration  5426ms) [ OK ]
Time: 1396 seconds

After:
ns4 MPTCP -> ns3 (10.0.3.2:10108      ) MPTCP   (duration  5417ms) [ OK ]
ns4 MPTCP -> ns3 (10.0.3.2:10109      ) TCP     (duration  5427ms) [ OK ]
ns4 TCP   -> ns3 (10.0.3.2:10110      ) MPTCP   (duration  5422ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10111) MPTCP   (duration  5415ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10112) TCP     (duration  5422ms) [ OK ]
ns4 TCP   -> ns3 (dead:beef:3::2:10113) MPTCP   (duration  5423ms) [ OK ]
Time: 296 seconds

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/mptcp/protocol.c | 123 ++++++++++++++++++++++++++++++++++++++++---
 net/mptcp/protocol.h |   7 +++
 net/mptcp/subflow.c  |   5 +-
 3 files changed, 127 insertions(+), 8 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 28ec26d97f96..fa137a9c42d1 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -179,13 +179,6 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk,
 		return false;
 	}
 
-	if (!(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
-		int rcvbuf = max(ssk->sk_rcvbuf, sk->sk_rcvbuf);
-
-		if (rcvbuf > sk->sk_rcvbuf)
-			sk->sk_rcvbuf = rcvbuf;
-	}
-
 	tp = tcp_sk(ssk);
 	do {
 		u32 map_remaining, offset;
@@ -916,6 +909,100 @@ static int __mptcp_recvmsg_mskq(struct mptcp_sock *msk,
 	return copied;
 }
 
+/* receive buffer autotuning.  See tcp_rcv_space_adjust for more information.
+ *
+ * Only difference: Use highest rtt estimate of the subflows in use.
+ */
+static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)
+{
+	struct mptcp_subflow_context *subflow;
+	struct sock *sk = (struct sock *)msk;
+	u32 time, advmss = 1;
+	u64 rtt_us, mstamp;
+
+	sock_owned_by_me(sk);
+
+	if (copied <= 0)
+		return;
+
+	msk->rcvq_space.copied += copied;
+
+	mstamp = div_u64(tcp_clock_ns(), NSEC_PER_USEC);
+	time = tcp_stamp_us_delta(mstamp, msk->rcvq_space.time);
+
+	rtt_us = msk->rcvq_space.rtt_us;
+	if (rtt_us && time < (rtt_us >> 3))
+		return;
+
+	rtt_us = 0;
+	mptcp_for_each_subflow(msk, subflow) {
+		const struct tcp_sock *tp;
+		u64 sf_rtt_us;
+		u32 sf_advmss;
+
+		tp = tcp_sk(mptcp_subflow_tcp_sock(subflow));
+
+		sf_rtt_us = READ_ONCE(tp->rcv_rtt_est.rtt_us);
+		sf_advmss = READ_ONCE(tp->advmss);
+
+		rtt_us = max(sf_rtt_us, rtt_us);
+		advmss = max(sf_advmss, advmss);
+	}
+
+	msk->rcvq_space.rtt_us = rtt_us;
+	if (time < (rtt_us >> 3) || rtt_us == 0)
+		return;
+
+	if (msk->rcvq_space.copied <= msk->rcvq_space.space)
+		goto new_measure;
+
+	if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf &&
+	    !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
+		int rcvmem, rcvbuf;
+		u64 rcvwin, grow;
+
+		rcvwin = ((u64)msk->rcvq_space.copied << 1) + 16 * advmss;
+
+		grow = rcvwin * (msk->rcvq_space.copied - msk->rcvq_space.space);
+
+		do_div(grow, msk->rcvq_space.space);
+		rcvwin += (grow << 1);
+
+		rcvmem = SKB_TRUESIZE(advmss + MAX_TCP_HEADER);
+		while (tcp_win_from_space(sk, rcvmem) < advmss)
+			rcvmem += 128;
+
+		do_div(rcvwin, advmss);
+		rcvbuf = min_t(u64, rcvwin * rcvmem,
+			       sock_net(sk)->ipv4.sysctl_tcp_rmem[2]);
+
+		if (rcvbuf > sk->sk_rcvbuf) {
+			u32 window_clamp;
+
+			window_clamp = tcp_win_from_space(sk, rcvbuf);
+			WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
+
+			/* Make subflows follow along.  If we do not do this, we
+			 * get drops at subflow level if skbs can't be moved to
+			 * the mptcp rx queue fast enough (announced rcv_win can
+			 * exceed ssk->sk_rcvbuf).
+			 */
+			mptcp_for_each_subflow(msk, subflow) {
+				struct sock *ssk;
+
+				ssk = mptcp_subflow_tcp_sock(subflow);
+				WRITE_ONCE(ssk->sk_rcvbuf, rcvbuf);
+				tcp_sk(ssk)->window_clamp = window_clamp;
+			}
+		}
+	}
+
+	msk->rcvq_space.space = msk->rcvq_space.copied;
+new_measure:
+	msk->rcvq_space.copied = 0;
+	msk->rcvq_space.time = mstamp;
+}
+
 static bool __mptcp_move_skbs(struct mptcp_sock *msk)
 {
 	unsigned int moved = 0;
@@ -1028,6 +1115,8 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 		set_bit(MPTCP_DATA_READY, &msk->flags);
 	}
 out_err:
+	mptcp_rcv_space_adjust(msk, copied);
+
 	release_sock(sk);
 	return copied;
 }
@@ -1241,6 +1330,7 @@ static int mptcp_init_sock(struct sock *sk)
 		return ret;
 
 	sk_sockets_allocated_inc(sk);
+	sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1];
 	sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[2];
 
 	return 0;
@@ -1423,6 +1513,22 @@ struct sock *mptcp_sk_clone(const struct sock *sk,
 	return nsk;
 }
 
+void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk)
+{
+	const struct tcp_sock *tp = tcp_sk(ssk);
+
+	msk->rcvq_space.copied = 0;
+	msk->rcvq_space.rtt_us = 0;
+
+	msk->rcvq_space.time = tp->tcp_mstamp;
+
+	/* initial rcv_space offering made to peer */
+	msk->rcvq_space.space = min_t(u32, tp->rcv_wnd,
+				      TCP_INIT_CWND * tp->advmss);
+	if (msk->rcvq_space.space == 0)
+		msk->rcvq_space.space = TCP_INIT_CWND * TCP_MSS_DEFAULT;
+}
+
 static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 				 bool kern)
 {
@@ -1471,6 +1577,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		list_add(&subflow->node, &msk->conn_list);
 		inet_sk_state_store(newsk, TCP_ESTABLISHED);
 
+		mptcp_rcv_space_init(msk, ssk);
 		bh_unlock_sock(new_mptcp_sock);
 
 		__MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPCAPABLEPASSIVEACK);
@@ -1631,6 +1738,8 @@ void mptcp_finish_connect(struct sock *ssk)
 	atomic64_set(&msk->snd_una, msk->write_seq);
 
 	mptcp_pm_new_connection(msk, 0);
+
+	mptcp_rcv_space_init(msk, ssk);
 }
 
 static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 1d05d9841b5c..a6412ff0fddb 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -209,6 +209,12 @@ struct mptcp_sock {
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
 	struct sock	*first;
 	struct mptcp_pm_data	pm;
+	struct {
+		u32	space;	/* bytes copied in last measurement window */
+		u32	copied; /* bytes copied in this measurement window */
+		u64	time;	/* start time of measurement window */
+		u64	rtt_us; /* last maximum rtt of subflows */
+	} rcvq_space;
 };
 
 #define mptcp_for_each_subflow(__msk, __subflow)			\
@@ -369,6 +375,7 @@ void mptcp_get_options(const struct sk_buff *skb,
 		       struct mptcp_options_received *mp_opt);
 
 void mptcp_finish_connect(struct sock *sk);
+void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk);
 void mptcp_data_ready(struct sock *sk, struct sock *ssk);
 bool mptcp_finish_join(struct sock *sk);
 void mptcp_data_acked(struct sock *sk);
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 548f9e347ff5..0e65bc07363c 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -253,8 +253,10 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 		pr_fallback(mptcp_sk(subflow->conn));
 	}
 
-	if (mptcp_check_fallback(sk))
+	if (mptcp_check_fallback(sk)) {
+		mptcp_rcv_space_init(mptcp_sk(parent), sk);
 		return;
+	}
 
 	if (subflow->mp_capable) {
 		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
@@ -1130,6 +1132,7 @@ static void subflow_state_change(struct sock *sk)
 
 	if (subflow_simultaneous_connect(sk)) {
 		mptcp_do_fallback(sk);
+		mptcp_rcv_space_init(mptcp_sk(parent), sk);
 		pr_fallback(mptcp_sk(parent));
 		subflow->conn_finished = 1;
 		if (inet_sk_state_load(parent) == TCP_SYN_SENT) {
-- 
2.26.2

Re: [PATCH net-next 1/2] selftests: mptcp: add option to specify size of file to transfer

[Matthieu Baerts]

Hi Florian,

On 30/06/2020 21:24, Florian Westphal wrote:
> The script generates two random files that are then sent via tcp and
> mptcp connections.
> 
> In order to compare throughput over consecutive runs add an option
> to provide the file size on the command line: "-f 128000".
> 
> Also add an option, -t, to enable tcp tests. This is useful to
> compare throughput of mptcp connections and tcp connections.
> 
> Example: run tests with a 4mb file size, 300ms delay 0.01% loss,
> default gso/tso/gro settings and with large write/blocking io:
> 
> mptcp_connect.sh -t -f $((4 * 1024 * 1024)) -d 300 -l 0.01%  -r 0 -e "" -m mmap
> 
> Signed-off-by: Florian Westphal <fw@strlen.de>

Thank you for the patch! The new '-t' option was also very helpful when 
investigating the issue 6 on Github!

Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>

Cheers,
Matt
--
Tessares | Belgium | Hybrid Access Solutions
www.tessares.net

Re: [PATCH net-next 2/2] mptcp: add receive buffer auto-tuning

[Matthieu Baerts]

Hi Florian,

On 30/06/2020 21:24, Florian Westphal wrote:
> When mptcp is used, userspace doesn't read from the tcp (subflow)
> socket but from the parent (mptcp) socket receive queue.
> 
> skbs are moved from the subflow socket to the mptcp rx queue either from
> 'data_ready' callback (if mptcp socket can be locked), a work queue, or
> the socket receive function.
> 
> This means tcp_rcv_space_adjust() is never called and thus no receive
> buffer size auto-tuning is done.
> 
> An earlier (not merged) patch added tcp_rcv_space_adjust() calls to the
> function that moves skbs from subflow to mptcp socket.
> While this enabled autotuning, it also meant tuning was done even if
> userspace was reading the mptcp socket very slowly.
> 
> This adds mptcp_rcv_space_adjust() and calls it after userspace has
> read data from the mptcp socket rx queue.
> 
> Its very similar to tcp_rcv_space_adjust, with two differences:
> 
> 1. The rtt estimate is the largest one observed on a subflow
> 2. The rcvbuf size and window clamp of all subflows is adjusted
>     to the mptcp-level rcvbuf.
> 
> Otherwise, we get spurious drops at tcp (subflow) socket level if
> the skbs are not moved to the mptcp socket fast enough.
> 
> Before:
> time mptcp_connect.sh -t -f $((4*1024*1024)) -d 300 -l 0.01% -r 0 -e "" -m mmap
> [..]
> ns4 MPTCP -> ns3 (10.0.3.2:10108      ) MPTCP   (duration 40823ms) [ OK ]
> ns4 MPTCP -> ns3 (10.0.3.2:10109      ) TCP     (duration 23119ms) [ OK ]
> ns4 TCP   -> ns3 (10.0.3.2:10110      ) MPTCP   (duration  5421ms) [ OK ]
> ns4 MPTCP -> ns3 (dead:beef:3::2:10111) MPTCP   (duration 41446ms) [ OK ]
> ns4 MPTCP -> ns3 (dead:beef:3::2:10112) TCP     (duration 23427ms) [ OK ]
> ns4 TCP   -> ns3 (dead:beef:3::2:10113) MPTCP   (duration  5426ms) [ OK ]
> Time: 1396 seconds
> 
> After:
> ns4 MPTCP -> ns3 (10.0.3.2:10108      ) MPTCP   (duration  5417ms) [ OK ]
> ns4 MPTCP -> ns3 (10.0.3.2:10109      ) TCP     (duration  5427ms) [ OK ]
> ns4 TCP   -> ns3 (10.0.3.2:10110      ) MPTCP   (duration  5422ms) [ OK ]
> ns4 MPTCP -> ns3 (dead:beef:3::2:10111) MPTCP   (duration  5415ms) [ OK ]
> ns4 MPTCP -> ns3 (dead:beef:3::2:10112) TCP     (duration  5422ms) [ OK ]
> ns4 TCP   -> ns3 (dead:beef:3::2:10113) MPTCP   (duration  5423ms) [ OK ]
> Time: 296 seconds
> 
> Signed-off-by: Florian Westphal <fw@strlen.de>

Thank you for adding this very useful feature!

Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>

Cheers,
Matt
-- 
Tessares | Belgium | Hybrid Access Solutions
www.tessares.net

Re: [PATCH net-next 0/2] mptcp: add receive buffer auto-tuning

[David Miller]

From: Florian Westphal <fw@strlen.de>
Date: Tue, 30 Jun 2020 21:24:43 +0200

> First patch extends the test script to allow for reproducible results.
> Second patch adds receive auto-tuning.  Its based on what TCP is doing,
> only difference is that we use the largest RTT of any of the subflows
> and that we will update all subflows with the new value.
> 
> Else, we get spurious packet drops because the mptcp work queue might
> not be able to move packets from subflow socket to master socket
> fast enough.  Without the adjustment, TCP may drop the packets because
> the subflow socket is over its rcvbuffer limit.

Series applied, thanks Florian.