[PATCH net-next v3 0/3] Optimizing performance in short-lived scenarios

[PATCH net-next v3 0/3] Optimizing performance in short-lived scenarios

[D. Wythe]

From: "D. Wythe" <alibuda@linux.alibaba.com>

This patch set aims to optimizing performance of SMC in short-lived
links scenarios, which is quite unsatisfactory right now.

In our benchmark, we test it with follow scripts:

./wrk -c 10000 -t 4 -H 'Connection: Close' -d 20 http://smc-server

Current performance figures like that:

Running 20s test @ http://11.213.45.6
  4 threads and 10000 connections
  4956 requests in 20.06s, 3.24MB read
  Socket errors: connect 0, read 0, write 672, timeout 0
Requests/sec:    247.07
Transfer/sec:    165.28KB

There are many reasons for this phenomenon, this patch set doesn't
solve it all though, but it can be well alleviated with it in.

Patch 1/3  (Make smc_tcp_listen_work() independent) :

Separate smc_tcp_listen_work() from smc_listen_work(), make them
independent of each other, the busy SMC handshake can not affect new TCP
connections visit any more. Avoid discarding a large number of TCP
connections after being overstock, which is undoubtedly raise the
connection establishment time.

Patch 2/3 (Limits SMC backlog connections):

Since patch 1 has separated smc_tcp_listen_work() from
smc_listen_work(), an unrestricted TCP accept have come into being. This
patch try to put a limit on SMC backlog connections refers to
implementation of TCP.

Patch 3/3 (Fallback when SMC handshake workqueue congested):

Considering the complexity of SMC handshake right now, in short-lived
links scenarios, this may not be the main scenario of SMC though, it's
performance is still quite poor. This Patch try to provide auto fallback
case when SMC handshake workqueue congested, which is the sign of SMC
handshake stacking in our opinion.

Of course, it's optional.

After this patch set, performance figures like that:

Running 20s test @ http://11.213.45.6
  4 threads and 10000 connections
  693253 requests in 20.10s, 452.88MB read
Requests/sec:  34488.13
Transfer/sec:     22.53MB

That's a quite well performance improvement, about to 6 to 7 times in my
environment.
---
changelog:
v2 -> v1:
- fix compile warning
- fix invalid dependencies in kconfig
v3 -> v2:
- correct spelling mistakes 
- fix useless variable declare
---
D. Wythe (3):
  net/smc: Make smc_tcp_listen_work() independent
  net/smc: Limits backlog connections
  net/smc: Fallback when handshake workqueue congested

 include/linux/tcp.h  |  1 +
 net/ipv4/tcp_input.c |  3 +-
 net/smc/Kconfig      | 12 ++++++++
 net/smc/af_smc.c     | 78 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 net/smc/smc.h        |  4 +++
 5 files changed, 95 insertions(+), 3 deletions(-)

-- 
1.8.3.1

[PATCH net-next v3 1/3] net/smc: Make smc_tcp_listen_work() independent

[D. Wythe]

From: "D. Wythe" <alibuda@linux.alibaba.com>

In multithread and 10K connections benchmark, the backend TCP connection
established very slowly, and lots of TCP connections stay in SYN_SENT
state.

Client: smc_run wrk -c 10000 -t 4 http://server

the netstate of server host shows like:
    145042 times the listen queue of a socket overflowed
    145042 SYNs to LISTEN sockets dropped

One reason of this issue is that, since the smc_tcp_listen_work() shared
the same workqueue (smc_hs_wq) with smc_listen_work(), while the
smc_listen_work() do blocking wait for smc connection established. Once
the workqueue became congested, it's will block the accept() from TCP
listen.

This patch creates a independent workqueue(smc_tcp_ls_wq) for
smc_tcp_listen_work(), separate it from smc_listen_work(), which is
quite acceptable considering that smc_tcp_listen_work() runs very fast.

Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
---
 net/smc/af_smc.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index d5ea62b..1b40304 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -59,6 +59,7 @@
 						 * creation on client
 						 */
 
+struct workqueue_struct	*smc_tcp_ls_wq;	/* wq for tcp listen work */
 struct workqueue_struct	*smc_hs_wq;	/* wq for handshake work */
 struct workqueue_struct	*smc_close_wq;	/* wq for close work */
 
@@ -2124,7 +2125,7 @@ static void smc_clcsock_data_ready(struct sock *listen_clcsock)
 	lsmc->clcsk_data_ready(listen_clcsock);
 	if (lsmc->sk.sk_state == SMC_LISTEN) {
 		sock_hold(&lsmc->sk); /* sock_put in smc_tcp_listen_work() */
-		if (!queue_work(smc_hs_wq, &lsmc->tcp_listen_work))
+		if (!queue_work(smc_tcp_ls_wq, &lsmc->tcp_listen_work))
 			sock_put(&lsmc->sk);
 	}
 }
@@ -2919,9 +2920,14 @@ static int __init smc_init(void)
 		goto out_nl;
 
 	rc = -ENOMEM;
+
+	smc_tcp_ls_wq = alloc_workqueue("smc_tcp_ls_wq", 0, 0);
+	if (!smc_tcp_ls_wq)
+		goto out_pnet;
+
 	smc_hs_wq = alloc_workqueue("smc_hs_wq", 0, 0);
 	if (!smc_hs_wq)
-		goto out_pnet;
+		goto out_alloc_tcp_ls_wq;
 
 	smc_close_wq = alloc_workqueue("smc_close_wq", 0, 0);
 	if (!smc_close_wq)
@@ -2992,6 +2998,8 @@ static int __init smc_init(void)
 	destroy_workqueue(smc_close_wq);
 out_alloc_hs_wq:
 	destroy_workqueue(smc_hs_wq);
+out_alloc_tcp_ls_wq:
+	destroy_workqueue(smc_tcp_ls_wq);
 out_pnet:
 	smc_pnet_exit();
 out_nl:
@@ -3010,6 +3018,7 @@ static void __exit smc_exit(void)
 	smc_core_exit();
 	smc_ib_unregister_client();
 	destroy_workqueue(smc_close_wq);
+	destroy_workqueue(smc_tcp_ls_wq);
 	destroy_workqueue(smc_hs_wq);
 	proto_unregister(&smc_proto6);
 	proto_unregister(&smc_proto);
-- 
1.8.3.1

[PATCH net-next v3 2/3] net/smc: Limits backlog connections

[D. Wythe]

From: "D. Wythe" <alibuda@linux.alibaba.com>

Current implementation does not handling backlog semantics, one
potential risk is that server will be flooded by infinite amount
connections, even if client was SMC-incapable.

This patch works to put a limit on backlog connections, referring to the
TCP implementation, we divides SMC connections into two categories:

1. Half SMC connection, which includes all TCP established while SMC not
connections.

2. Full SMC connection, which includes all SMC established connections.

For half SMC connection, since all half SMC connections starts with TCP
established, we can achieve our goal by put a limit before TCP
established. Refer to the implementation of TCP, this limits will based
on not only the half SMC connections but also the full connections,
which is also a constraint on full SMC connections.

For full SMC connections, although we know exactly where it starts, it's
quite hard to put a limit before it. The easiest way is to block wait
before receive SMC confirm CLC message, while it's under protection by
smc_server_lgr_pending, a global lock, which leads this limit to the
entire host instead of a single listen socket. Another way is to drop
the full connections, but considering the cast of SMC connections, we
prefer to keep full SMC connections.

Even so, the limits of full SMC connections still exists, see commits
about half SMC connection below.

After this patch, the limits of backend connection shows like:

For SMC:

1. Client with SMC-capability can makes 2 * backlog full SMC connections
   or 1 * backlog half SMC connections and 1 * backlog full SMC
   connections at most.

2. Client without SMC-capability can only makes 1 * backlog half TCP
   connections and 1 * backlog full TCP connections.

Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
---
 net/smc/af_smc.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 net/smc/smc.h    |  4 ++++
 2 files changed, 47 insertions(+)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 1b40304..66a0e64 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -73,6 +73,34 @@ static void smc_set_keepalive(struct sock *sk, int val)
 	smc->clcsock->sk->sk_prot->keepalive(smc->clcsock->sk, val);
 }
 
+static struct sock *smc_tcp_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
+					  struct request_sock *req,
+					  struct dst_entry *dst,
+					  struct request_sock *req_unhash,
+					  bool *own_req)
+{
+	struct smc_sock *smc;
+
+	smc = (struct smc_sock *)((uintptr_t)sk->sk_user_data & ~SK_USER_DATA_NOCOPY);
+
+	if (READ_ONCE(sk->sk_ack_backlog) + atomic_read(&smc->smc_pendings) >
+				sk->sk_max_ack_backlog)
+		goto drop;
+
+	if (sk_acceptq_is_full(&smc->sk)) {
+		NET_INC_STATS(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
+		goto drop;
+	}
+
+	/* passthrough to origin syn recv sock fct */
+	return smc->ori_af_ops->syn_recv_sock(sk, skb, req, dst, req_unhash, own_req);
+
+drop:
+	dst_release(dst);
+	tcp_listendrop(sk);
+	return NULL;
+}
+
 static struct smc_hashinfo smc_v4_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock),
 };
@@ -1491,6 +1519,9 @@ static void smc_listen_out(struct smc_sock *new_smc)
 	struct smc_sock *lsmc = new_smc->listen_smc;
 	struct sock *newsmcsk = &new_smc->sk;
 
+	if (tcp_sk(new_smc->clcsock->sk)->syn_smc)
+		atomic_dec(&lsmc->smc_pendings);
+
 	if (lsmc->sk.sk_state == SMC_LISTEN) {
 		lock_sock_nested(&lsmc->sk, SINGLE_DEPTH_NESTING);
 		smc_accept_enqueue(&lsmc->sk, newsmcsk);
@@ -2096,6 +2127,9 @@ static void smc_tcp_listen_work(struct work_struct *work)
 		if (!new_smc)
 			continue;
 
+		if (tcp_sk(new_smc->clcsock->sk)->syn_smc)
+			atomic_inc(&lsmc->smc_pendings);
+
 		new_smc->listen_smc = lsmc;
 		new_smc->use_fallback = lsmc->use_fallback;
 		new_smc->fallback_rsn = lsmc->fallback_rsn;
@@ -2163,6 +2197,15 @@ static int smc_listen(struct socket *sock, int backlog)
 	smc->clcsock->sk->sk_data_ready = smc_clcsock_data_ready;
 	smc->clcsock->sk->sk_user_data =
 		(void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
+
+	/* save origin ops */
+	smc->ori_af_ops = inet_csk(smc->clcsock->sk)->icsk_af_ops;
+
+	smc->af_ops = *smc->ori_af_ops;
+	smc->af_ops.syn_recv_sock = smc_tcp_syn_recv_sock;
+
+	inet_csk(smc->clcsock->sk)->icsk_af_ops = &smc->af_ops;
+
 	rc = kernel_listen(smc->clcsock, backlog);
 	if (rc) {
 		smc->clcsock->sk->sk_data_ready = smc->clcsk_data_ready;
diff --git a/net/smc/smc.h b/net/smc/smc.h
index 3d0b8e3..2b9f443 100644
--- a/net/smc/smc.h
+++ b/net/smc/smc.h
@@ -240,6 +240,10 @@ struct smc_sock {				/* smc sock container */
 	bool			use_fallback;	/* fallback to tcp */
 	int			fallback_rsn;	/* reason for fallback */
 	u32			peer_diagnosis; /* decline reason from peer */
+	atomic_t                smc_pendings;   /* pending smc connections */
+	struct inet_connection_sock_af_ops		af_ops;
+	const struct inet_connection_sock_af_ops	*ori_af_ops;
+						/* origin af ops */
 	int			sockopt_defer_accept;
 						/* sockopt TCP_DEFER_ACCEPT
 						 * value
-- 
1.8.3.1

[PATCH net-next v3 3/3] net/smc: Fallback when handshake workqueue congested

[D. Wythe]

From: "D. Wythe" <alibuda@linux.alibaba.com>

This patch intends to provide a mechanism to allow automatic fallback to
TCP according to the pressure of SMC handshake process. At present,
frequent visits will cause the incoming connections to be backlogged in
SMC handshake queue, raise the connections established time. Which is
quite unacceptable for those applications who base on short lived
connections.

It should be optional for applications that don't care about connection
established time. For now, this patch only provides the switch at the
compile time.

There are two ways to implement this mechanism:

1. Fallback when TCP established.
2. Fallback before TCP established.

In the first way, we need to wait and receive CLC messages that the
client will potentially send, and then actively reply with a decline
message, in a sense, which is also a sort of SMC handshake, affect the
connections established time on its way.

In the second way, the only problem is that we need to inject SMC logic
into TCP when it is about to reply the incoming SYN, since we already do
that, it's seems not a problem anymore. And advantage is obvious, few
additional processes are required to complete the fallback.

This patch use the second way.

Link: https://lore.kernel.org/all/1641301961-59331-1-git-send-email-alibuda@linux.alibaba.com/
Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
---
 include/linux/tcp.h  |  1 +
 net/ipv4/tcp_input.c |  3 ++-
 net/smc/Kconfig      | 12 ++++++++++++
 net/smc/af_smc.c     | 22 ++++++++++++++++++++++
 4 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 78b91bb..1c4ae5d 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -394,6 +394,7 @@ struct tcp_sock {
 	bool	is_mptcp;
 #endif
 #if IS_ENABLED(CONFIG_SMC)
+	bool	(*smc_in_limited)(const struct sock *sk);
 	bool	syn_smc;	/* SYN includes SMC */
 #endif
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index dc49a3d..9890de9 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6701,7 +6701,8 @@ static void tcp_openreq_init(struct request_sock *req,
 	ireq->ir_num = ntohs(tcp_hdr(skb)->dest);
 	ireq->ir_mark = inet_request_mark(sk, skb);
 #if IS_ENABLED(CONFIG_SMC)
-	ireq->smc_ok = rx_opt->smc_ok;
+	ireq->smc_ok = rx_opt->smc_ok && !(tcp_sk(sk)->smc_in_limited &&
+			tcp_sk(sk)->smc_in_limited(sk));
 #endif
 }
 
diff --git a/net/smc/Kconfig b/net/smc/Kconfig
index 1ab3c5a..a4e1713 100644
--- a/net/smc/Kconfig
+++ b/net/smc/Kconfig
@@ -19,3 +19,15 @@ config SMC_DIAG
 	  smcss.
 
 	  if unsure, say Y.
+
+if SMC
+
+config SMC_AUTO_FALLBACK
+	bool "SMC: automatic fallback to TCP"
+	default y
+	help
+	  Allow automatic fallback to TCP accroding to the pressure of SMC-R
+	  handshake process.
+
+	  If that's not what you except or unsure, say N.
+endif
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 66a0e64..49b8a29 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -101,6 +101,24 @@ static struct sock *smc_tcp_syn_recv_sock(const struct sock *sk, struct sk_buff
 	return NULL;
 }
 
+#if IS_ENABLED(CONFIG_SMC_AUTO_FALLBACK)
+static bool smc_is_in_limited(const struct sock *sk)
+{
+	const struct smc_sock *smc;
+
+	smc = (const struct smc_sock *)
+		((uintptr_t)sk->sk_user_data & ~SK_USER_DATA_NOCOPY);
+
+	if (!smc)
+		return true;
+
+	if (workqueue_congested(WORK_CPU_UNBOUND, smc_hs_wq))
+		return true;
+
+	return false;
+}
+#endif
+
 static struct smc_hashinfo smc_v4_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock),
 };
@@ -2206,6 +2224,10 @@ static int smc_listen(struct socket *sock, int backlog)
 
 	inet_csk(smc->clcsock->sk)->icsk_af_ops = &smc->af_ops;
 
+#if IS_ENABLED(CONFIG_SMC_AUTO_FALLBACK)
+	tcp_sk(smc->clcsock->sk)->smc_in_limited = smc_is_in_limited;
+#endif
+
 	rc = kernel_listen(smc->clcsock, backlog);
 	if (rc) {
 		smc->clcsock->sk->sk_data_ready = smc->clcsk_data_ready;
-- 
1.8.3.1

[RFC PATCH] net/smc: smc_tcp_ls_wq can be static

[kernel test robot]

net/smc/af_smc.c:62:25: warning: symbol 'smc_tcp_ls_wq' was not declared. Should it be static?

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: kernel test robot <lkp@intel.com>
---
 af_smc.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 05b88cbadf3d1..4969ac8029a98 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -59,7 +59,7 @@ static DEFINE_MUTEX(smc_client_lgr_pending);	/* serialize link group
 						 * creation on client
 						 */
 
-struct workqueue_struct	*smc_tcp_ls_wq;	/* wq for tcp listen work */
+static struct workqueue_struct	*smc_tcp_ls_wq;	/* wq for tcp listen work */
 struct workqueue_struct	*smc_hs_wq;	/* wq for handshake work */
 struct workqueue_struct	*smc_close_wq;	/* wq for close work */
 

[RFC PATCH] net/smc: smc_tcp_ls_wq can be static

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 827 bytes --]

net/smc/af_smc.c:62:25: warning: symbol 'smc_tcp_ls_wq' was not declared. Should it be static?

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: kernel test robot <lkp@intel.com>
---
 af_smc.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 05b88cbadf3d1..4969ac8029a98 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -59,7 +59,7 @@ static DEFINE_MUTEX(smc_client_lgr_pending);	/* serialize link group
 						 * creation on client
 						 */
 
-struct workqueue_struct	*smc_tcp_ls_wq;	/* wq for tcp listen work */
+static struct workqueue_struct	*smc_tcp_ls_wq;	/* wq for tcp listen work */
 struct workqueue_struct	*smc_hs_wq;	/* wq for handshake work */
 struct workqueue_struct	*smc_close_wq;	/* wq for close work */
 

Re: [PATCH net-next v3 1/3] net/smc: Make smc_tcp_listen_work() independent

[kernel test robot]

Hi Wythe",

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on net-next/master]

url:    https://github.com/0day-ci/linux/commits/D-Wythe/Optimizing-performance-in-short-lived-scenarios/20220205-143638
base:   https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git c78b8b20e34920231eda02fb40c7aca7d88be837
config: x86_64-randconfig-s021 (https://download.01.org/0day-ci/archive/20220205/202202052019.I6XO0mge-lkp@intel.com/config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
reproduce:
        # apt-get install sparse
        # sparse version: v0.6.4-dirty
        # https://github.com/0day-ci/linux/commit/c750e361a42f2d32b4e041edfd0b51d9020a936c
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review D-Wythe/Optimizing-performance-in-short-lived-scenarios/20220205-143638
        git checkout c750e361a42f2d32b4e041edfd0b51d9020a936c
        # save the config file to linux build tree
        mkdir build_dir
        make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=x86_64 SHELL=/bin/bash net/smc/

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)
>> net/smc/af_smc.c:62:25: sparse: sparse: symbol 'smc_tcp_ls_wq' was not declared. Should it be static?

Please review and possibly fold the followup patch.

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

Re: [PATCH net-next v3 1/3] net/smc: Make smc_tcp_listen_work() independent

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 1585 bytes --]

Hi Wythe",

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on net-next/master]

url:    https://github.com/0day-ci/linux/commits/D-Wythe/Optimizing-performance-in-short-lived-scenarios/20220205-143638
base:   https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git c78b8b20e34920231eda02fb40c7aca7d88be837
config: x86_64-randconfig-s021 (https://download.01.org/0day-ci/archive/20220205/202202052019.I6XO0mge-lkp(a)intel.com/config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
reproduce:
        # apt-get install sparse
        # sparse version: v0.6.4-dirty
        # https://github.com/0day-ci/linux/commit/c750e361a42f2d32b4e041edfd0b51d9020a936c
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review D-Wythe/Optimizing-performance-in-short-lived-scenarios/20220205-143638
        git checkout c750e361a42f2d32b4e041edfd0b51d9020a936c
        # save the config file to linux build tree
        mkdir build_dir
        make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=x86_64 SHELL=/bin/bash net/smc/

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)
>> net/smc/af_smc.c:62:25: sparse: sparse: symbol 'smc_tcp_ls_wq' was not declared. Should it be static?

Please review and possibly fold the followup patch.

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org