diff mbox series

[net-next,v2,1/4] net/smc: Add SMC statistics support

Message ID 20210616145258.2381446-2-kgraul@linux.ibm.com (mailing list archive)
State Accepted
Commit e0e4b8fa533858532f1b9ea9c6a4660d09beb37a
Delegated to: Netdev Maintainers
Headers show
Series net/smc: Add SMC statistic support | expand

Checks

Context Check Description
netdev/cover_letter success Link
netdev/fixes_present success Link
netdev/patch_count success Link
netdev/tree_selection success Clearly marked for net-next
netdev/subject_prefix success Link
netdev/cc_maintainers success CCed 5 of 5 maintainers
netdev/source_inline success Was 0 now: 0
netdev/verify_signedoff success Link
netdev/module_param success Was 0 now: 0
netdev/build_32bit success Errors and warnings before: 0 this patch: 0
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/verify_fixes success Link
netdev/checkpatch warning WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
netdev/build_allmodconfig_warn success Errors and warnings before: 0 this patch: 0
netdev/header_inline success Link

Commit Message

Karsten Graul June 16, 2021, 2:52 p.m. UTC
From: Guvenc Gulce <guvenc@linux.ibm.com>

Add the ability to collect SMC statistics information. Per-cpu
variables are used to collect the statistic information for better
performance and for reducing concurrency pitfalls. The code that is
collecting statistic data is implemented in macros to increase code
reuse and readability.

Signed-off-by: Guvenc Gulce <guvenc@linux.ibm.com>
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
---
 net/smc/Makefile    |   2 +-
 net/smc/af_smc.c    |  89 ++++++++++++----
 net/smc/smc_core.c  |  13 ++-
 net/smc/smc_rx.c    |   8 ++
 net/smc/smc_stats.c |  35 ++++++
 net/smc/smc_stats.h | 253 ++++++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_tx.c    |  16 ++-
 7 files changed, 395 insertions(+), 21 deletions(-)
 create mode 100644 net/smc/smc_stats.c
 create mode 100644 net/smc/smc_stats.h
diff mbox series

Patch

diff --git a/net/smc/Makefile b/net/smc/Makefile
index 77e54fe42b1c..99a0186cba5b 100644
--- a/net/smc/Makefile
+++ b/net/smc/Makefile
@@ -2,4 +2,4 @@ 
 obj-$(CONFIG_SMC)	+= smc.o
 obj-$(CONFIG_SMC_DIAG)	+= smc_diag.o
 smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o
-smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o
+smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 5eff7cccceff..efeaed384769 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -49,6 +49,7 @@ 
 #include "smc_tx.h"
 #include "smc_rx.h"
 #include "smc_close.h"
+#include "smc_stats.h"
 
 static DEFINE_MUTEX(smc_server_lgr_pending);	/* serialize link group
 						 * creation on server
@@ -508,9 +509,42 @@  static void smc_link_save_peer_info(struct smc_link *link,
 	link->peer_mtu = clc->r0.qp_mtu;
 }
 
-static void smc_switch_to_fallback(struct smc_sock *smc)
+static void smc_stat_inc_fback_rsn_cnt(struct smc_sock *smc,
+				       struct smc_stats_fback *fback_arr)
+{
+	int cnt;
+
+	for (cnt = 0; cnt < SMC_MAX_FBACK_RSN_CNT; cnt++) {
+		if (fback_arr[cnt].fback_code == smc->fallback_rsn) {
+			fback_arr[cnt].count++;
+			break;
+		}
+		if (!fback_arr[cnt].fback_code) {
+			fback_arr[cnt].fback_code = smc->fallback_rsn;
+			fback_arr[cnt].count++;
+			break;
+		}
+	}
+}
+
+static void smc_stat_fallback(struct smc_sock *smc)
+{
+	mutex_lock(&smc_stat_fback_rsn);
+	if (smc->listen_smc) {
+		smc_stat_inc_fback_rsn_cnt(smc, fback_rsn.srv);
+		fback_rsn.srv_fback_cnt++;
+	} else {
+		smc_stat_inc_fback_rsn_cnt(smc, fback_rsn.clnt);
+		fback_rsn.clnt_fback_cnt++;
+	}
+	mutex_unlock(&smc_stat_fback_rsn);
+}
+
+static void smc_switch_to_fallback(struct smc_sock *smc, int reason_code)
 {
 	smc->use_fallback = true;
+	smc->fallback_rsn = reason_code;
+	smc_stat_fallback(smc);
 	if (smc->sk.sk_socket && smc->sk.sk_socket->file) {
 		smc->clcsock->file = smc->sk.sk_socket->file;
 		smc->clcsock->file->private_data = smc->clcsock;
@@ -522,8 +556,7 @@  static void smc_switch_to_fallback(struct smc_sock *smc)
 /* fall back during connect */
 static int smc_connect_fallback(struct smc_sock *smc, int reason_code)
 {
-	smc_switch_to_fallback(smc);
-	smc->fallback_rsn = reason_code;
+	smc_switch_to_fallback(smc, reason_code);
 	smc_copy_sock_settings_to_clc(smc);
 	smc->connect_nonblock = 0;
 	if (smc->sk.sk_state == SMC_INIT)
@@ -538,6 +571,7 @@  static int smc_connect_decline_fallback(struct smc_sock *smc, int reason_code,
 	int rc;
 
 	if (reason_code < 0) { /* error, fallback is not possible */
+		this_cpu_inc(smc_stats->clnt_hshake_err_cnt);
 		if (smc->sk.sk_state == SMC_INIT)
 			sock_put(&smc->sk); /* passive closing */
 		return reason_code;
@@ -545,6 +579,7 @@  static int smc_connect_decline_fallback(struct smc_sock *smc, int reason_code,
 	if (reason_code != SMC_CLC_DECL_PEERDECL) {
 		rc = smc_clc_send_decline(smc, reason_code, version);
 		if (rc < 0) {
+			this_cpu_inc(smc_stats->clnt_hshake_err_cnt);
 			if (smc->sk.sk_state == SMC_INIT)
 				sock_put(&smc->sk); /* passive closing */
 			return rc;
@@ -992,6 +1027,7 @@  static int __smc_connect(struct smc_sock *smc)
 	if (rc)
 		goto vlan_cleanup;
 
+	SMC_STAT_CLNT_SUCC_INC(aclc);
 	smc_connect_ism_vlan_cleanup(smc, ini);
 	kfree(buf);
 	kfree(ini);
@@ -1308,6 +1344,7 @@  static void smc_listen_out_err(struct smc_sock *new_smc)
 {
 	struct sock *newsmcsk = &new_smc->sk;
 
+	this_cpu_inc(smc_stats->srv_hshake_err_cnt);
 	if (newsmcsk->sk_state == SMC_INIT)
 		sock_put(&new_smc->sk); /* passive closing */
 	newsmcsk->sk_state = SMC_CLOSED;
@@ -1325,8 +1362,7 @@  static void smc_listen_decline(struct smc_sock *new_smc, int reason_code,
 		smc_listen_out_err(new_smc);
 		return;
 	}
-	smc_switch_to_fallback(new_smc);
-	new_smc->fallback_rsn = reason_code;
+	smc_switch_to_fallback(new_smc, reason_code);
 	if (reason_code && reason_code != SMC_CLC_DECL_PEERDECL) {
 		if (smc_clc_send_decline(new_smc, reason_code, version) < 0) {
 			smc_listen_out_err(new_smc);
@@ -1699,8 +1735,7 @@  static void smc_listen_work(struct work_struct *work)
 
 	/* check if peer is smc capable */
 	if (!tcp_sk(newclcsock->sk)->syn_smc) {
-		smc_switch_to_fallback(new_smc);
-		new_smc->fallback_rsn = SMC_CLC_DECL_PEERNOSMC;
+		smc_switch_to_fallback(new_smc, SMC_CLC_DECL_PEERNOSMC);
 		smc_listen_out_connected(new_smc);
 		return;
 	}
@@ -1778,6 +1813,7 @@  static void smc_listen_work(struct work_struct *work)
 	}
 	smc_conn_save_peer_info(new_smc, cclc);
 	smc_listen_out_connected(new_smc);
+	SMC_STAT_SERV_SUCC_INC(ini);
 	goto out_free;
 
 out_unlock:
@@ -1984,18 +2020,19 @@  static int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 
 	if (msg->msg_flags & MSG_FASTOPEN) {
 		if (sk->sk_state == SMC_INIT && !smc->connect_nonblock) {
-			smc_switch_to_fallback(smc);
-			smc->fallback_rsn = SMC_CLC_DECL_OPTUNSUPP;
+			smc_switch_to_fallback(smc, SMC_CLC_DECL_OPTUNSUPP);
 		} else {
 			rc = -EINVAL;
 			goto out;
 		}
 	}
 
-	if (smc->use_fallback)
+	if (smc->use_fallback) {
 		rc = smc->clcsock->ops->sendmsg(smc->clcsock, msg, len);
-	else
+	} else {
 		rc = smc_tx_sendmsg(smc, msg, len);
+		SMC_STAT_TX_PAYLOAD(smc, len, rc);
+	}
 out:
 	release_sock(sk);
 	return rc;
@@ -2030,6 +2067,7 @@  static int smc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	} else {
 		msg->msg_namelen = 0;
 		rc = smc_rx_recvmsg(smc, msg, NULL, len, flags);
+		SMC_STAT_RX_PAYLOAD(smc, rc, rc);
 	}
 
 out:
@@ -2194,8 +2232,7 @@  static int smc_setsockopt(struct socket *sock, int level, int optname,
 	case TCP_FASTOPEN_NO_COOKIE:
 		/* option not supported by SMC */
 		if (sk->sk_state == SMC_INIT && !smc->connect_nonblock) {
-			smc_switch_to_fallback(smc);
-			smc->fallback_rsn = SMC_CLC_DECL_OPTUNSUPP;
+			smc_switch_to_fallback(smc, SMC_CLC_DECL_OPTUNSUPP);
 		} else {
 			rc = -EINVAL;
 		}
@@ -2204,18 +2241,22 @@  static int smc_setsockopt(struct socket *sock, int level, int optname,
 		if (sk->sk_state != SMC_INIT &&
 		    sk->sk_state != SMC_LISTEN &&
 		    sk->sk_state != SMC_CLOSED) {
-			if (val)
+			if (val) {
+				SMC_STAT_INC(!smc->conn.lnk, ndly_cnt);
 				mod_delayed_work(smc->conn.lgr->tx_wq,
 						 &smc->conn.tx_work, 0);
+			}
 		}
 		break;
 	case TCP_CORK:
 		if (sk->sk_state != SMC_INIT &&
 		    sk->sk_state != SMC_LISTEN &&
 		    sk->sk_state != SMC_CLOSED) {
-			if (!val)
+			if (!val) {
+				SMC_STAT_INC(!smc->conn.lnk, cork_cnt);
 				mod_delayed_work(smc->conn.lgr->tx_wq,
 						 &smc->conn.tx_work, 0);
+			}
 		}
 		break;
 	case TCP_DEFER_ACCEPT:
@@ -2338,11 +2379,13 @@  static ssize_t smc_sendpage(struct socket *sock, struct page *page,
 		goto out;
 	}
 	release_sock(sk);
-	if (smc->use_fallback)
+	if (smc->use_fallback) {
 		rc = kernel_sendpage(smc->clcsock, page, offset,
 				     size, flags);
-	else
+	} else {
+		SMC_STAT_INC(!smc->conn.lnk, sendpage_cnt);
 		rc = sock_no_sendpage(sock, page, offset, size, flags);
+	}
 
 out:
 	return rc;
@@ -2391,6 +2434,7 @@  static ssize_t smc_splice_read(struct socket *sock, loff_t *ppos,
 			flags = MSG_DONTWAIT;
 		else
 			flags = 0;
+		SMC_STAT_INC(!smc->conn.lnk, splice_cnt);
 		rc = smc_rx_recvmsg(smc, NULL, pipe, len, flags);
 	}
 out:
@@ -2514,10 +2558,16 @@  static int __init smc_init(void)
 	if (!smc_close_wq)
 		goto out_alloc_hs_wq;
 
+	rc = smc_stats_init();
+	if (rc) {
+		pr_err("%s: smc_stats_init fails with %d\n", __func__, rc);
+		goto out_alloc_wqs;
+	}
+
 	rc = smc_core_init();
 	if (rc) {
 		pr_err("%s: smc_core_init fails with %d\n", __func__, rc);
-		goto out_alloc_wqs;
+		goto out_smc_stat;
 	}
 
 	rc = smc_llc_init();
@@ -2569,6 +2619,8 @@  static int __init smc_init(void)
 	proto_unregister(&smc_proto);
 out_core:
 	smc_core_exit();
+out_smc_stat:
+	smc_stats_exit();
 out_alloc_wqs:
 	destroy_workqueue(smc_close_wq);
 out_alloc_hs_wq:
@@ -2591,6 +2643,7 @@  static void __exit smc_exit(void)
 	smc_ib_unregister_client();
 	destroy_workqueue(smc_close_wq);
 	destroy_workqueue(smc_hs_wq);
+	smc_stats_exit();
 	proto_unregister(&smc_proto6);
 	proto_unregister(&smc_proto);
 	smc_pnet_exit();
diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c
index 317bc2c90fab..d69f58f670a1 100644
--- a/net/smc/smc_core.c
+++ b/net/smc/smc_core.c
@@ -33,6 +33,7 @@ 
 #include "smc_close.h"
 #include "smc_ism.h"
 #include "smc_netlink.h"
+#include "smc_stats.h"
 
 #define SMC_LGR_NUM_INCR		256
 #define SMC_LGR_FREE_DELAY_SERV		(600 * HZ)
@@ -2029,6 +2030,7 @@  static int __smc_buf_create(struct smc_sock *smc, bool is_smcd, bool is_rmb)
 	struct smc_link_group *lgr = conn->lgr;
 	struct list_head *buf_list;
 	int bufsize, bufsize_short;
+	bool is_dgraded = false;
 	struct mutex *lock;	/* lock buffer list */
 	int sk_buf_size;
 
@@ -2056,6 +2058,8 @@  static int __smc_buf_create(struct smc_sock *smc, bool is_smcd, bool is_rmb)
 		/* check for reusable slot in the link group */
 		buf_desc = smc_buf_get_slot(bufsize_short, lock, buf_list);
 		if (buf_desc) {
+			SMC_STAT_RMB_SIZE(is_smcd, is_rmb, bufsize);
+			SMC_STAT_BUF_REUSE(is_smcd, is_rmb);
 			memset(buf_desc->cpu_addr, 0, bufsize);
 			break; /* found reusable slot */
 		}
@@ -2067,9 +2071,16 @@  static int __smc_buf_create(struct smc_sock *smc, bool is_smcd, bool is_rmb)
 
 		if (PTR_ERR(buf_desc) == -ENOMEM)
 			break;
-		if (IS_ERR(buf_desc))
+		if (IS_ERR(buf_desc)) {
+			if (!is_dgraded) {
+				is_dgraded = true;
+				SMC_STAT_RMB_DOWNGRADED(is_smcd, is_rmb);
+			}
 			continue;
+		}
 
+		SMC_STAT_RMB_ALLOC(is_smcd, is_rmb);
+		SMC_STAT_RMB_SIZE(is_smcd, is_rmb, bufsize);
 		buf_desc->used = 1;
 		mutex_lock(lock);
 		list_add(&buf_desc->list, buf_list);
diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c
index fcfac59f8b72..ce1ae39923b1 100644
--- a/net/smc/smc_rx.c
+++ b/net/smc/smc_rx.c
@@ -21,6 +21,7 @@ 
 #include "smc_cdc.h"
 #include "smc_tx.h" /* smc_tx_consumer_update() */
 #include "smc_rx.h"
+#include "smc_stats.h"
 
 /* callback implementation to wakeup consumers blocked with smc_rx_wait().
  * indirectly called by smc_cdc_msg_recv_action().
@@ -227,6 +228,7 @@  static int smc_rx_recv_urg(struct smc_sock *smc, struct msghdr *msg, int len,
 	    conn->urg_state == SMC_URG_READ)
 		return -EINVAL;
 
+	SMC_STAT_INC(!conn->lnk, urg_data_cnt);
 	if (conn->urg_state == SMC_URG_VALID) {
 		if (!(flags & MSG_PEEK))
 			smc->conn.urg_state = SMC_URG_READ;
@@ -303,6 +305,12 @@  int smc_rx_recvmsg(struct smc_sock *smc, struct msghdr *msg,
 	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
 	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
 
+	readable = atomic_read(&conn->bytes_to_rcv);
+	if (readable >= conn->rmb_desc->len)
+		SMC_STAT_RMB_RX_FULL(!conn->lnk);
+
+	if (len < readable)
+		SMC_STAT_RMB_RX_SIZE_SMALL(!conn->lnk);
 	/* we currently use 1 RMBE per RMB, so RMBE == RMB base addr */
 	rcvbuf_base = conn->rx_off + conn->rmb_desc->cpu_addr;
 
diff --git a/net/smc/smc_stats.c b/net/smc/smc_stats.c
new file mode 100644
index 000000000000..76e938388520
--- /dev/null
+++ b/net/smc/smc_stats.c
@@ -0,0 +1,35 @@ 
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * SMC statistics netlink routines
+ *
+ * Copyright IBM Corp. 2021
+ *
+ * Author(s):  Guvenc Gulce
+ */
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/percpu.h>
+#include <linux/ctype.h>
+#include "smc_stats.h"
+
+/* serialize fallback reason statistic gathering */
+DEFINE_MUTEX(smc_stat_fback_rsn);
+struct smc_stats __percpu *smc_stats;	/* per cpu counters for SMC */
+struct smc_stats_reason fback_rsn;
+
+int __init smc_stats_init(void)
+{
+	memset(&fback_rsn, 0, sizeof(fback_rsn));
+	smc_stats = alloc_percpu(struct smc_stats);
+	if (!smc_stats)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void smc_stats_exit(void)
+{
+	free_percpu(smc_stats);
+}
diff --git a/net/smc/smc_stats.h b/net/smc/smc_stats.h
new file mode 100644
index 000000000000..928372114cf1
--- /dev/null
+++ b/net/smc/smc_stats.h
@@ -0,0 +1,253 @@ 
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Macros for SMC statistics
+ *
+ * Copyright IBM Corp. 2021
+ *
+ * Author(s):  Guvenc Gulce
+ */
+
+#ifndef NET_SMC_SMC_STATS_H_
+#define NET_SMC_SMC_STATS_H_
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/percpu.h>
+#include <linux/ctype.h>
+#include <linux/smc.h>
+
+#include "smc_clc.h"
+
+#define SMC_MAX_FBACK_RSN_CNT 30
+
+extern struct smc_stats __percpu *smc_stats;	/* per cpu counters for SMC */
+extern struct smc_stats_reason fback_rsn;
+extern struct mutex smc_stat_fback_rsn;
+
+enum {
+	SMC_BUF_8K,
+	SMC_BUF_16K,
+	SMC_BUF_32K,
+	SMC_BUF_64K,
+	SMC_BUF_128K,
+	SMC_BUF_256K,
+	SMC_BUF_512K,
+	SMC_BUF_1024K,
+	SMC_BUF_G_1024K,
+	SMC_BUF_MAX,
+};
+
+struct smc_stats_fback {
+	int	fback_code;
+	u16	count;
+};
+
+struct smc_stats_reason {
+	struct	smc_stats_fback srv[SMC_MAX_FBACK_RSN_CNT];
+	struct	smc_stats_fback clnt[SMC_MAX_FBACK_RSN_CNT];
+	u64			srv_fback_cnt;
+	u64			clnt_fback_cnt;
+};
+
+struct smc_stats_rmbcnt {
+	u64	buf_size_small_peer_cnt;
+	u64	buf_size_small_cnt;
+	u64	buf_full_peer_cnt;
+	u64	buf_full_cnt;
+	u64	reuse_cnt;
+	u64	alloc_cnt;
+	u64	dgrade_cnt;
+};
+
+struct smc_stats_memsize {
+	u64	buf[SMC_BUF_MAX];
+};
+
+struct smc_stats_tech {
+	struct smc_stats_memsize tx_rmbsize;
+	struct smc_stats_memsize rx_rmbsize;
+	struct smc_stats_memsize tx_pd;
+	struct smc_stats_memsize rx_pd;
+	struct smc_stats_rmbcnt rmb_tx;
+	struct smc_stats_rmbcnt rmb_rx;
+	u64			clnt_v1_succ_cnt;
+	u64			clnt_v2_succ_cnt;
+	u64			srv_v1_succ_cnt;
+	u64			srv_v2_succ_cnt;
+	u64			sendpage_cnt;
+	u64			urg_data_cnt;
+	u64			splice_cnt;
+	u64			cork_cnt;
+	u64			ndly_cnt;
+	u64			rx_bytes;
+	u64			tx_bytes;
+	u64			rx_cnt;
+	u64			tx_cnt;
+};
+
+struct smc_stats {
+	struct smc_stats_tech	smc[2];
+	u64			clnt_hshake_err_cnt;
+	u64			srv_hshake_err_cnt;
+};
+
+#define SMC_STAT_PAYLOAD_SUB(_tech, key, _len, _rc) \
+do { \
+	typeof(_tech) t = (_tech); \
+	typeof(_len) l = (_len); \
+	int _pos = fls64((l) >> 13); \
+	typeof(_rc) r = (_rc); \
+	int m = SMC_BUF_MAX - 1; \
+	this_cpu_inc((*smc_stats).smc[t].key ## _cnt); \
+	if (r <= 0) \
+		break; \
+	_pos = (_pos < m) ? ((l == 1 << (_pos + 12)) ? _pos - 1 : _pos) : m; \
+	this_cpu_inc((*smc_stats).smc[t].key ## _pd.buf[_pos]); \
+	this_cpu_add((*smc_stats).smc[t].key ## _bytes, r); \
+} \
+while (0)
+
+#define SMC_STAT_TX_PAYLOAD(_smc, length, rcode) \
+do { \
+	typeof(_smc) __smc = _smc; \
+	typeof(length) _len = (length); \
+	typeof(rcode) _rc = (rcode); \
+	bool is_smcd = !__smc->conn.lnk; \
+	if (is_smcd) \
+		SMC_STAT_PAYLOAD_SUB(SMC_TYPE_D, tx, _len, _rc); \
+	else \
+		SMC_STAT_PAYLOAD_SUB(SMC_TYPE_R, tx, _len, _rc); \
+} \
+while (0)
+
+#define SMC_STAT_RX_PAYLOAD(_smc, length, rcode) \
+do { \
+	typeof(_smc) __smc = _smc; \
+	typeof(length) _len = (length); \
+	typeof(rcode) _rc = (rcode); \
+	bool is_smcd = !__smc->conn.lnk; \
+	if (is_smcd) \
+		SMC_STAT_PAYLOAD_SUB(SMC_TYPE_D, rx, _len, _rc); \
+	else \
+		SMC_STAT_PAYLOAD_SUB(SMC_TYPE_R, rx, _len, _rc); \
+} \
+while (0)
+
+#define SMC_STAT_RMB_SIZE_SUB(_tech, k, _len) \
+do { \
+	typeof(_len) _l = (_len); \
+	typeof(_tech) t = (_tech); \
+	int _pos = fls((_l) >> 13); \
+	int m = SMC_BUF_MAX - 1; \
+	_pos = (_pos < m) ? ((_l == 1 << (_pos + 12)) ? _pos - 1 : _pos) : m; \
+	this_cpu_inc((*smc_stats).smc[t].k ## _rmbsize.buf[_pos]); \
+} \
+while (0)
+
+#define SMC_STAT_RMB_SUB(type, t, key) \
+	this_cpu_inc((*smc_stats).smc[t].rmb ## _ ## key.type ## _cnt)
+
+#define SMC_STAT_RMB_SIZE(_is_smcd, _is_rx, _len) \
+do { \
+	typeof(_is_smcd) is_d = (_is_smcd); \
+	typeof(_is_rx) is_r = (_is_rx); \
+	typeof(_len) l = (_len); \
+	if ((is_d) && (is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(SMC_TYPE_D, rx, l); \
+	if ((is_d) && !(is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(SMC_TYPE_D, tx, l); \
+	if (!(is_d) && (is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(SMC_TYPE_R, rx, l); \
+	if (!(is_d) && !(is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(SMC_TYPE_R, tx, l); \
+} \
+while (0)
+
+#define SMC_STAT_RMB(type, _is_smcd, _is_rx) \
+do { \
+	typeof(_is_smcd) is_d = (_is_smcd); \
+	typeof(_is_rx) is_r = (_is_rx); \
+	if ((is_d) && (is_r)) \
+		SMC_STAT_RMB_SUB(type, SMC_TYPE_D, rx); \
+	if ((is_d) && !(is_r)) \
+		SMC_STAT_RMB_SUB(type, SMC_TYPE_D, tx); \
+	if (!(is_d) && (is_r)) \
+		SMC_STAT_RMB_SUB(type, SMC_TYPE_R, rx); \
+	if (!(is_d) && !(is_r)) \
+		SMC_STAT_RMB_SUB(type, SMC_TYPE_R, tx); \
+} \
+while (0)
+
+#define SMC_STAT_BUF_REUSE(is_smcd, is_rx) \
+	SMC_STAT_RMB(reuse, is_smcd, is_rx)
+
+#define SMC_STAT_RMB_ALLOC(is_smcd, is_rx) \
+	SMC_STAT_RMB(alloc, is_smcd, is_rx)
+
+#define SMC_STAT_RMB_DOWNGRADED(is_smcd, is_rx) \
+	SMC_STAT_RMB(dgrade, is_smcd, is_rx)
+
+#define SMC_STAT_RMB_TX_PEER_FULL(is_smcd) \
+	SMC_STAT_RMB(buf_full_peer, is_smcd, false)
+
+#define SMC_STAT_RMB_TX_FULL(is_smcd) \
+	SMC_STAT_RMB(buf_full, is_smcd, false)
+
+#define SMC_STAT_RMB_TX_PEER_SIZE_SMALL(is_smcd) \
+	SMC_STAT_RMB(buf_size_small_peer, is_smcd, false)
+
+#define SMC_STAT_RMB_TX_SIZE_SMALL(is_smcd) \
+	SMC_STAT_RMB(buf_size_small, is_smcd, false)
+
+#define SMC_STAT_RMB_RX_SIZE_SMALL(is_smcd) \
+	SMC_STAT_RMB(buf_size_small, is_smcd, true)
+
+#define SMC_STAT_RMB_RX_FULL(is_smcd) \
+	SMC_STAT_RMB(buf_full, is_smcd, true)
+
+#define SMC_STAT_INC(is_smcd, type) \
+do { \
+	if ((is_smcd)) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_D].type); \
+	else \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_R].type); \
+} \
+while (0)
+
+#define SMC_STAT_CLNT_SUCC_INC(_aclc) \
+do { \
+	typeof(_aclc) acl = (_aclc); \
+	bool is_v2 = (acl->hdr.version == SMC_V2); \
+	bool is_smcd = (acl->hdr.typev1 == SMC_TYPE_D); \
+	if (is_v2 && is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_D].clnt_v2_succ_cnt); \
+	else if (is_v2 && !is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_R].clnt_v2_succ_cnt); \
+	else if (!is_v2 && is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_D].clnt_v1_succ_cnt); \
+	else if (!is_v2 && !is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_R].clnt_v1_succ_cnt); \
+} \
+while (0)
+
+#define SMC_STAT_SERV_SUCC_INC(_ini) \
+do { \
+	typeof(_ini) i = (_ini); \
+	bool is_v2 = (i->smcd_version & SMC_V2); \
+	bool is_smcd = (i->is_smcd); \
+	if (is_v2 && is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_D].srv_v2_succ_cnt); \
+	else if (is_v2 && !is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_R].srv_v2_succ_cnt); \
+	else if (!is_v2 && is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_D].srv_v1_succ_cnt); \
+	else if (!is_v2 && !is_smcd) \
+		this_cpu_inc(smc_stats->smc[SMC_TYPE_R].srv_v1_succ_cnt); \
+} \
+while (0)
+
+int smc_stats_init(void) __init;
+void smc_stats_exit(void);
+
+#endif /* NET_SMC_SMC_STATS_H_ */
diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c
index 4532c16bf85e..a043544d715f 100644
--- a/net/smc/smc_tx.c
+++ b/net/smc/smc_tx.c
@@ -27,6 +27,7 @@ 
 #include "smc_close.h"
 #include "smc_ism.h"
 #include "smc_tx.h"
+#include "smc_stats.h"
 
 #define SMC_TX_WORK_DELAY	0
 #define SMC_TX_CORK_DELAY	(HZ >> 2)	/* 250 ms */
@@ -45,6 +46,8 @@  static void smc_tx_write_space(struct sock *sk)
 
 	/* similar to sk_stream_write_space */
 	if (atomic_read(&smc->conn.sndbuf_space) && sock) {
+		if (test_bit(SOCK_NOSPACE, &sock->flags))
+			SMC_STAT_RMB_TX_FULL(!smc->conn.lnk);
 		clear_bit(SOCK_NOSPACE, &sock->flags);
 		rcu_read_lock();
 		wq = rcu_dereference(sk->sk_wq);
@@ -151,6 +154,15 @@  int smc_tx_sendmsg(struct smc_sock *smc, struct msghdr *msg, size_t len)
 		goto out_err;
 	}
 
+	if (len > conn->sndbuf_desc->len)
+		SMC_STAT_RMB_TX_SIZE_SMALL(!conn->lnk);
+
+	if (len > conn->peer_rmbe_size)
+		SMC_STAT_RMB_TX_PEER_SIZE_SMALL(!conn->lnk);
+
+	if (msg->msg_flags & MSG_OOB)
+		SMC_STAT_INC(!conn->lnk, urg_data_cnt);
+
 	while (msg_data_left(msg)) {
 		if (sk->sk_state == SMC_INIT)
 			return -ENOTCONN;
@@ -419,8 +431,10 @@  static int smc_tx_rdma_writes(struct smc_connection *conn,
 	/* destination: RMBE */
 	/* cf. snd_wnd */
 	rmbespace = atomic_read(&conn->peer_rmbe_space);
-	if (rmbespace <= 0)
+	if (rmbespace <= 0) {
+		SMC_STAT_RMB_TX_PEER_FULL(!conn->lnk);
 		return 0;
+	}
 	smc_curs_copy(&prod, &conn->local_tx_ctrl.prod, conn);
 	smc_curs_copy(&cons, &conn->local_rx_ctrl.cons, conn);