Resguardo inicial

linux-7.2.3/net/smc/Kconfig

@@ -0,0 +1,31 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config SMC
+	tristate "SMC socket protocol family"
+	depends on INET && INFINIBAND && DIBS
+	help
+	  SMC-R provides a "sockets over RDMA" solution making use of
+	  RDMA over Converged Ethernet (RoCE) technology to upgrade
+	  AF_INET TCP connections transparently.
+	  The Linux implementation of the SMC-R solution is designed as
+	  a separate socket family SMC.
+
+	  Select this option if you want to run SMC socket applications
+
+config SMC_DIAG
+	tristate "SMC: socket monitoring interface"
+	depends on SMC
+	help
+	  Support for SMC socket monitoring interface used by tools such as
+	  smcss.
+
+	  if unsure, say Y.
+
+config SMC_HS_CTRL_BPF
+	bool "Generic eBPF hook for SMC handshake flow"
+	depends on SMC && BPF_JIT && BPF_SYSCALL
+	default y
+	help
+	  SMC_HS_CTRL_BPF enables support to register generic eBPF hook for SMC
+	  handshake flow, which offer much greater flexibility in modifying the behavior
+	  of the SMC protocol stack compared to a complete kernel-based approach. Select
+	  this option if you want filtring the handshake process via eBPF programs.

linux-7.2.3/net/smc/Makefile

@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
+ccflags-y += -I$(src)
+obj-$(CONFIG_SMC)	+= smc.o
+obj-$(CONFIG_SMC_DIAG)	+= smc_diag.o
+smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o
+smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o
+smc-y += smc_tracepoint.o smc_inet.o
+smc-$(CONFIG_SYSCTL) += smc_sysctl.o
+smc-$(CONFIG_SMC_HS_CTRL_BPF) += smc_hs_bpf.o

linux-7.2.3/net/smc/smc.h

@@ -0,0 +1,434 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for the SMC module (socket related)
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+#ifndef __SMC_H
+#define __SMC_H
+
+#include <linux/socket.h>
+#include <linux/types.h>
+#include <linux/compiler.h> /* __aligned */
+#include <net/genetlink.h>
+#include <net/sock.h>
+
+#include "smc_ib.h"
+
+#define SMC_V1		1		/* SMC version V1 */
+#define SMC_V2		2		/* SMC version V2 */
+
+#define SMC_RELEASE_0 0
+#define SMC_RELEASE_1 1
+#define SMC_RELEASE	SMC_RELEASE_1 /* the latest release version */
+
+#define SMCPROTO_SMC		0	/* SMC protocol, IPv4 */
+#define SMCPROTO_SMC6		1	/* SMC protocol, IPv6 */
+
+#define SMC_AUTOCORKING_DEFAULT_SIZE	0x10000	/* 64K by default */
+
+extern struct proto smc_proto;
+extern struct proto smc_proto6;
+
+extern struct smc_hashinfo smc_v4_hashinfo;
+extern struct smc_hashinfo smc_v6_hashinfo;
+
+int smc_hash_sk(struct sock *sk);
+void smc_unhash_sk(struct sock *sk);
+void smc_release_cb(struct sock *sk);
+
+int smc_release(struct socket *sock);
+int smc_bind(struct socket *sock, struct sockaddr_unsized *uaddr,
+	     int addr_len);
+int smc_connect(struct socket *sock, struct sockaddr_unsized *addr,
+		int alen, int flags);
+int smc_accept(struct socket *sock, struct socket *new_sock,
+	       struct proto_accept_arg *arg);
+int smc_getname(struct socket *sock, struct sockaddr *addr,
+		int peer);
+__poll_t smc_poll(struct file *file, struct socket *sock,
+		  poll_table *wait);
+int smc_ioctl(struct socket *sock, unsigned int cmd,
+	      unsigned long arg);
+int smc_listen(struct socket *sock, int backlog);
+int smc_shutdown(struct socket *sock, int how);
+int smc_setsockopt(struct socket *sock, int level, int optname,
+		   sockptr_t optval, unsigned int optlen);
+int smc_getsockopt(struct socket *sock, int level, int optname,
+		   char __user *optval, int __user *optlen);
+int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len);
+int smc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
+		int flags);
+ssize_t smc_splice_read(struct socket *sock, loff_t *ppos,
+			struct pipe_inode_info *pipe, size_t len,
+			unsigned int flags);
+
+/* smc sock initialization */
+void smc_sk_init(struct net *net, struct sock *sk, int protocol);
+/* clcsock initialization */
+int smc_create_clcsk(struct net *net, struct sock *sk, int family);
+
+#ifdef ATOMIC64_INIT
+#define KERNEL_HAS_ATOMIC64
+#endif
+
+enum smc_state {		/* possible states of an SMC socket */
+	SMC_ACTIVE	= 1,
+	SMC_INIT	= 2,
+	SMC_CLOSED	= 7,
+	SMC_LISTEN	= 10,
+	/* normal close */
+	SMC_PEERCLOSEWAIT1	= 20,
+	SMC_PEERCLOSEWAIT2	= 21,
+	SMC_APPFINCLOSEWAIT	= 24,
+	SMC_APPCLOSEWAIT1	= 22,
+	SMC_APPCLOSEWAIT2	= 23,
+	SMC_PEERFINCLOSEWAIT	= 25,
+	/* abnormal close */
+	SMC_PEERABORTWAIT	= 26,
+	SMC_PROCESSABORT	= 27,
+};
+
+enum smc_supplemental_features {
+	SMC_SPF_EMULATED_ISM_DEV	= 0,
+};
+
+#define SMC_FEATURE_MASK \
+	(BIT(SMC_SPF_EMULATED_ISM_DEV))
+
+struct smc_link_group;
+
+struct smc_wr_rx_hdr {	/* common prefix part of LLC and CDC to demultiplex */
+	union {
+		u8 type;
+#if defined(__BIG_ENDIAN_BITFIELD)
+		struct {
+			u8 llc_version:4,
+			   llc_type:4;
+		};
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+		struct {
+			u8 llc_type:4,
+			   llc_version:4;
+		};
+#endif
+	};
+} __aligned(1);
+
+struct smc_cdc_conn_state_flags {
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8	peer_done_writing : 1;	/* Sending done indicator */
+	u8	peer_conn_closed : 1;	/* Peer connection closed indicator */
+	u8	peer_conn_abort : 1;	/* Abnormal close indicator */
+	u8	reserved : 5;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8	reserved : 5;
+	u8	peer_conn_abort : 1;
+	u8	peer_conn_closed : 1;
+	u8	peer_done_writing : 1;
+#endif
+};
+
+struct smc_cdc_producer_flags {
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8	write_blocked : 1;	/* Writing Blocked, no rx buf space */
+	u8	urg_data_pending : 1;	/* Urgent Data Pending */
+	u8	urg_data_present : 1;	/* Urgent Data Present */
+	u8	cons_curs_upd_req : 1;	/* cursor update requested */
+	u8	failover_validation : 1;/* message replay due to failover */
+	u8	reserved : 3;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8	reserved : 3;
+	u8	failover_validation : 1;
+	u8	cons_curs_upd_req : 1;
+	u8	urg_data_present : 1;
+	u8	urg_data_pending : 1;
+	u8	write_blocked : 1;
+#endif
+};
+
+/* in host byte order */
+union smc_host_cursor {	/* SMC cursor - an offset in an RMBE */
+	struct {
+		u16	reserved;
+		u16	wrap;		/* window wrap sequence number */
+		u32	count;		/* cursor (= offset) part */
+	};
+#ifdef KERNEL_HAS_ATOMIC64
+	atomic64_t		acurs;	/* for atomic processing */
+#else
+	u64			acurs;	/* for atomic processing */
+#endif
+} __aligned(8);
+
+/* in host byte order, except for flag bitfields in network byte order */
+struct smc_host_cdc_msg {		/* Connection Data Control message */
+	struct smc_wr_rx_hdr		common; /* .type = 0xFE */
+	u8				len;	/* length = 44 */
+	u16				seqno;	/* connection seq # */
+	u32				token;	/* alert_token */
+	union smc_host_cursor		prod;		/* producer cursor */
+	union smc_host_cursor		cons;		/* consumer cursor,
+							 * piggy backed "ack"
+							 */
+	struct smc_cdc_producer_flags	prod_flags;	/* conn. tx/rx status */
+	struct smc_cdc_conn_state_flags	conn_state_flags; /* peer conn. status*/
+	u8				reserved[18];
+} __aligned(8);
+
+enum smc_urg_state {
+	SMC_URG_VALID	= 1,			/* data present */
+	SMC_URG_NOTYET	= 2,			/* data pending */
+	SMC_URG_READ	= 3,			/* data was already read */
+};
+
+struct smc_mark_woken {
+	bool woken;
+	void *key;
+	wait_queue_entry_t wait_entry;
+};
+
+struct smc_connection {
+	struct rb_node		alert_node;
+	struct smc_link_group	*lgr;		/* link group of connection */
+	struct smc_link		*lnk;		/* assigned SMC-R link */
+	u32			alert_token_local; /* unique conn. id */
+	u8			peer_rmbe_idx;	/* from tcp handshake */
+	int			peer_rmbe_size;	/* size of peer rx buffer */
+	atomic_t		peer_rmbe_space;/* remaining free bytes in peer
+						 * rmbe
+						 */
+	int			rtoken_idx;	/* idx to peer RMB rkey/addr */
+
+	struct smc_buf_desc	*sndbuf_desc;	/* send buffer descriptor */
+	struct smc_buf_desc	*rmb_desc;	/* RMBE descriptor */
+	int                     rmbe_size_comp; /* compressed notation */
+	int			rmbe_update_limit;
+						/* lower limit for consumer
+						 * cursor update
+						 */
+
+	struct smc_host_cdc_msg	local_tx_ctrl;	/* host byte order staging
+						 * buffer for CDC msg send
+						 * .prod cf. TCP snd_nxt
+						 * .cons cf. TCP sends ack
+						 */
+	union smc_host_cursor	local_tx_ctrl_fin;
+						/* prod crsr - confirmed by peer
+						 */
+	union smc_host_cursor	tx_curs_prep;	/* tx - prepared data
+						 * snd_max..wmem_alloc
+						 */
+	union smc_host_cursor	tx_curs_sent;	/* tx - sent data
+						 * snd_nxt ?
+						 */
+	union smc_host_cursor	tx_curs_fin;	/* tx - confirmed by peer
+						 * snd-wnd-begin ?
+						 */
+	atomic_t		sndbuf_space;	/* remaining space in sndbuf */
+	u16			tx_cdc_seq;	/* sequence # for CDC send */
+	u16			tx_cdc_seq_fin;	/* sequence # - tx completed */
+	spinlock_t		send_lock;	/* protect wr_sends */
+	atomic_t		cdc_pend_tx_wr; /* number of pending tx CDC wqe
+						 * - inc when post wqe,
+						 * - dec on polled tx cqe
+						 */
+	wait_queue_head_t	cdc_pend_tx_wq; /* wakeup on no cdc_pend_tx_wr*/
+	struct delayed_work	tx_work;	/* retry of smc_cdc_msg_send */
+	u32			tx_off;		/* base offset in peer rmb */
+
+	struct smc_host_cdc_msg	local_rx_ctrl;	/* filled during event_handl.
+						 * .prod cf. TCP rcv_nxt
+						 * .cons cf. TCP snd_una
+						 */
+	union smc_host_cursor	rx_curs_confirmed; /* confirmed to peer
+						    * source of snd_una ?
+						    */
+	union smc_host_cursor	urg_curs;	/* points at urgent byte */
+	enum smc_urg_state	urg_state;
+	bool			urg_tx_pend;	/* urgent data staged */
+	bool			urg_rx_skip_pend;
+						/* indicate urgent oob data
+						 * read, but previous regular
+						 * data still pending
+						 */
+	char			urg_rx_byte;	/* urgent byte */
+	bool			tx_in_release_sock;
+						/* flush pending tx data in
+						 * sock release_cb()
+						 */
+	atomic_t		bytes_to_rcv;	/* arrived data,
+						 * not yet received
+						 */
+	atomic_t		splice_pending;	/* number of spliced bytes
+						 * pending processing
+						 */
+#ifndef KERNEL_HAS_ATOMIC64
+	spinlock_t		acurs_lock;	/* protect cursors */
+#endif
+	struct work_struct	close_work;	/* peer sent some closing */
+	struct work_struct	abort_work;	/* abort the connection */
+	struct tasklet_struct	rx_tsklet;	/* Receiver tasklet for SMC-D */
+	u8			rx_off;		/* receive offset:
+						 * 0 for SMC-R, 32 for SMC-D
+						 */
+	u64			peer_token;	/* SMC-D token of peer */
+	u8			killed : 1;	/* abnormal termination */
+	u8			freed : 1;	/* normal termination */
+	u8			out_of_sync : 1; /* out of sync with peer */
+};
+
+struct smc_sock {				/* smc sock container */
+	union {
+		struct sock		sk;
+		struct inet_sock	icsk_inet;
+	};
+	struct socket		*clcsock;	/* internal tcp socket */
+	void			(*clcsk_state_change)(struct sock *sk);
+						/* original stat_change fct. */
+	void			(*clcsk_data_ready)(struct sock *sk);
+						/* original data_ready fct. */
+	void			(*clcsk_write_space)(struct sock *sk);
+						/* original write_space fct. */
+	void			(*clcsk_error_report)(struct sock *sk);
+						/* original error_report fct. */
+	struct smc_connection	conn;		/* smc connection */
+	struct smc_sock		*listen_smc;	/* listen parent */
+	struct work_struct	connect_work;	/* handle non-blocking connect*/
+	struct work_struct	tcp_listen_work;/* handle tcp socket accepts */
+	struct work_struct	smc_listen_work;/* prepare new accept socket */
+	struct list_head	accept_q;	/* sockets to be accepted */
+	spinlock_t		accept_q_lock;	/* protects accept_q */
+	bool			limit_smc_hs;	/* put constraint on handshake */
+	bool			use_fallback;	/* fallback to tcp */
+	int			fallback_rsn;	/* reason for fallback */
+	u32			peer_diagnosis; /* decline reason from peer */
+	atomic_t                queued_smc_hs;  /* queued smc handshakes */
+	struct inet_connection_sock_af_ops		af_ops;
+	const struct inet_connection_sock_af_ops	*ori_af_ops;
+						/* original af ops */
+	int			sockopt_defer_accept;
+						/* sockopt TCP_DEFER_ACCEPT
+						 * value
+						 */
+	u8			wait_close_tx_prepared : 1;
+						/* shutdown wr or close
+						 * started, waiting for unsent
+						 * data to be sent
+						 */
+	u8			connect_nonblock : 1;
+						/* non-blocking connect in
+						 * flight
+						 */
+	struct mutex            clcsock_release_lock;
+						/* protects clcsock of a listen
+						 * socket
+						 * */
+};
+
+#define smc_sk(ptr) container_of_const(ptr, struct smc_sock, sk)
+
+static inline void smc_init_saved_callbacks(struct smc_sock *smc)
+{
+	smc->clcsk_state_change	= NULL;
+	smc->clcsk_data_ready	= NULL;
+	smc->clcsk_write_space	= NULL;
+	smc->clcsk_error_report	= NULL;
+}
+
+static inline struct smc_sock *smc_clcsock_user_data(const struct sock *clcsk)
+{
+	return (struct smc_sock *)
+	       ((uintptr_t)clcsk->sk_user_data & ~SK_USER_DATA_NOCOPY);
+}
+
+static inline struct smc_sock *smc_clcsock_user_data_rcu(const struct sock *clcsk)
+{
+	return (struct smc_sock *)rcu_dereference_sk_user_data(clcsk);
+}
+
+/* save target_cb in saved_cb, and replace target_cb with new_cb */
+static inline void smc_clcsock_replace_cb(void (**target_cb)(struct sock *),
+					  void (*new_cb)(struct sock *),
+					  void (**saved_cb)(struct sock *))
+{
+	/* only save once */
+	if (!*saved_cb)
+		*saved_cb = *target_cb;
+	*target_cb = new_cb;
+}
+
+/* restore target_cb to saved_cb, and reset saved_cb to NULL */
+static inline void smc_clcsock_restore_cb(void (**target_cb)(struct sock *),
+					  void (**saved_cb)(struct sock *))
+{
+	if (!*saved_cb)
+		return;
+	*target_cb = *saved_cb;
+	*saved_cb = NULL;
+}
+
+extern struct workqueue_struct	*smc_hs_wq;	/* wq for handshake work */
+extern struct workqueue_struct	*smc_close_wq;	/* wq for close work */
+
+#define SMC_SYSTEMID_LEN		8
+
+extern u8	local_systemid[SMC_SYSTEMID_LEN]; /* unique system identifier */
+
+#define ntohll(x) be64_to_cpu(x)
+#define htonll(x) cpu_to_be64(x)
+
+/* convert an u32 value into network byte order, store it into a 3 byte field */
+static inline void hton24(u8 *net, u32 host)
+{
+	__be32 t;
+
+	t = cpu_to_be32(host);
+	memcpy(net, ((u8 *)&t) + 1, 3);
+}
+
+/* convert a received 3 byte field into host byte order*/
+static inline u32 ntoh24(u8 *net)
+{
+	__be32 t = 0;
+
+	memcpy(((u8 *)&t) + 1, net, 3);
+	return be32_to_cpu(t);
+}
+
+#ifdef CONFIG_XFRM
+static inline bool using_ipsec(struct smc_sock *smc)
+{
+	return (smc->clcsock->sk->sk_policy[0] ||
+		smc->clcsock->sk->sk_policy[1]) ? true : false;
+}
+#else
+static inline bool using_ipsec(struct smc_sock *smc)
+{
+	return false;
+}
+#endif
+
+struct smc_gidlist;
+
+struct sock *smc_accept_dequeue(struct sock *parent, struct socket *new_sock);
+void smc_close_non_accepted(struct sock *sk);
+void smc_fill_gid_list(struct smc_link_group *lgr,
+		       struct smc_gidlist *gidlist,
+		       struct smc_ib_device *known_dev, u8 *known_gid);
+
+/* smc handshake limitation interface for netlink  */
+int smc_nl_dump_hs_limitation(struct sk_buff *skb, struct netlink_callback *cb);
+int smc_nl_enable_hs_limitation(struct sk_buff *skb, struct genl_info *info);
+int smc_nl_disable_hs_limitation(struct sk_buff *skb, struct genl_info *info);
+
+static inline void smc_sock_set_flag(struct sock *sk, enum sock_flags flag)
+{
+	set_bit(flag, &sk->sk_flags);
+}
+
+#endif	/* __SMC_H */

linux-7.2.3/net/smc/smc_cdc.c

@@ -0,0 +1,530 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Connection Data Control (CDC)
+ * handles flow control
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#include <linux/spinlock.h>
+
+#include "smc.h"
+#include "smc_wr.h"
+#include "smc_cdc.h"
+#include "smc_tx.h"
+#include "smc_rx.h"
+#include "smc_close.h"
+#include "smc_ism.h"
+
+/********************************** send *************************************/
+
+/* handler for send/transmission completion of a CDC msg */
+static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd,
+			       struct smc_link *link,
+			       enum ib_wc_status wc_status)
+{
+	struct smc_cdc_tx_pend *cdcpend = (struct smc_cdc_tx_pend *)pnd_snd;
+	struct smc_connection *conn = cdcpend->conn;
+	struct smc_buf_desc *sndbuf_desc;
+	struct smc_sock *smc;
+	int diff;
+
+	sndbuf_desc = conn->sndbuf_desc;
+	smc = container_of(conn, struct smc_sock, conn);
+	bh_lock_sock(&smc->sk);
+	if (!wc_status && sndbuf_desc) {
+		diff = smc_curs_diff(sndbuf_desc->len,
+				     &cdcpend->conn->tx_curs_fin,
+				     &cdcpend->cursor);
+		/* sndbuf_space is decreased in smc_sendmsg */
+		smp_mb__before_atomic();
+		atomic_add(diff, &cdcpend->conn->sndbuf_space);
+		/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
+		smp_mb__after_atomic();
+		smc_curs_copy(&conn->tx_curs_fin, &cdcpend->cursor, conn);
+		smc_curs_copy(&conn->local_tx_ctrl_fin, &cdcpend->p_cursor,
+			      conn);
+		conn->tx_cdc_seq_fin = cdcpend->ctrl_seq;
+	}
+
+	if (atomic_dec_and_test(&conn->cdc_pend_tx_wr)) {
+		/* If user owns the sock_lock, mark the connection need sending.
+		 * User context will later try to send when it release sock_lock
+		 * in smc_release_cb()
+		 */
+		if (sock_owned_by_user(&smc->sk))
+			conn->tx_in_release_sock = true;
+		else
+			smc_tx_pending(conn);
+
+		if (unlikely(wq_has_sleeper(&conn->cdc_pend_tx_wq)))
+			wake_up(&conn->cdc_pend_tx_wq);
+	}
+	WARN_ON(atomic_read(&conn->cdc_pend_tx_wr) < 0);
+
+	smc_tx_sndbuf_nonfull(smc);
+	bh_unlock_sock(&smc->sk);
+}
+
+int smc_cdc_get_free_slot(struct smc_connection *conn,
+			  struct smc_link *link,
+			  struct smc_wr_buf **wr_buf,
+			  struct smc_rdma_wr **wr_rdma_buf,
+			  struct smc_cdc_tx_pend **pend)
+{
+	int rc;
+
+	rc = smc_wr_tx_get_free_slot(link, smc_cdc_tx_handler, wr_buf,
+				     wr_rdma_buf,
+				     (struct smc_wr_tx_pend_priv **)pend);
+	if (conn->killed) {
+		/* abnormal termination */
+		if (!rc)
+			smc_wr_tx_put_slot(link,
+					   (struct smc_wr_tx_pend_priv *)(*pend));
+		rc = -EPIPE;
+	}
+	return rc;
+}
+
+static inline void smc_cdc_add_pending_send(struct smc_connection *conn,
+					    struct smc_cdc_tx_pend *pend)
+{
+	BUILD_BUG_ON_MSG(
+		sizeof(struct smc_cdc_msg) > SMC_WR_BUF_SIZE,
+		"must increase SMC_WR_BUF_SIZE to at least sizeof(struct smc_cdc_msg)");
+	BUILD_BUG_ON_MSG(
+		offsetofend(struct smc_cdc_msg, reserved) > SMC_WR_TX_SIZE,
+		"must adapt SMC_WR_TX_SIZE to sizeof(struct smc_cdc_msg); if not all smc_wr upper layer protocols use the same message size any more, must start to set link->wr_tx_sges[i].length on each individual smc_wr_tx_send()");
+	BUILD_BUG_ON_MSG(
+		sizeof(struct smc_cdc_tx_pend) > SMC_WR_TX_PEND_PRIV_SIZE,
+		"must increase SMC_WR_TX_PEND_PRIV_SIZE to at least sizeof(struct smc_cdc_tx_pend)");
+	pend->conn = conn;
+	pend->cursor = conn->tx_curs_sent;
+	pend->p_cursor = conn->local_tx_ctrl.prod;
+	pend->ctrl_seq = conn->tx_cdc_seq;
+}
+
+int smc_cdc_msg_send(struct smc_connection *conn,
+		     struct smc_wr_buf *wr_buf,
+		     struct smc_cdc_tx_pend *pend)
+{
+	struct smc_link *link = conn->lnk;
+	union smc_host_cursor cfed;
+	int rc;
+
+	smc_cdc_add_pending_send(conn, pend);
+
+	conn->tx_cdc_seq++;
+	conn->local_tx_ctrl.seqno = conn->tx_cdc_seq;
+	smc_host_msg_to_cdc((struct smc_cdc_msg *)wr_buf, conn, &cfed);
+
+	atomic_inc(&conn->cdc_pend_tx_wr);
+	smp_mb__after_atomic(); /* Make sure cdc_pend_tx_wr added before post */
+
+	rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend);
+	if (!rc) {
+		smc_curs_copy(&conn->rx_curs_confirmed, &cfed, conn);
+		conn->local_rx_ctrl.prod_flags.cons_curs_upd_req = 0;
+	} else {
+		conn->tx_cdc_seq--;
+		conn->local_tx_ctrl.seqno = conn->tx_cdc_seq;
+		atomic_dec(&conn->cdc_pend_tx_wr);
+	}
+
+	return rc;
+}
+
+/* send a validation msg indicating the move of a conn to an other QP link */
+int smcr_cdc_msg_send_validation(struct smc_connection *conn,
+				 struct smc_cdc_tx_pend *pend,
+				 struct smc_wr_buf *wr_buf)
+{
+	struct smc_host_cdc_msg *local = &conn->local_tx_ctrl;
+	struct smc_link *link = conn->lnk;
+	struct smc_cdc_msg *peer;
+	int rc;
+
+	peer = (struct smc_cdc_msg *)wr_buf;
+	peer->common.type = local->common.type;
+	peer->len = local->len;
+	peer->seqno = htons(conn->tx_cdc_seq_fin); /* seqno last compl. tx */
+	peer->token = htonl(local->token);
+	peer->prod_flags.failover_validation = 1;
+
+	/* We need to set pend->conn here to make sure smc_cdc_tx_handler()
+	 * can handle properly
+	 */
+	smc_cdc_add_pending_send(conn, pend);
+
+	atomic_inc(&conn->cdc_pend_tx_wr);
+	smp_mb__after_atomic(); /* Make sure cdc_pend_tx_wr added before post */
+
+	rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend);
+	if (unlikely(rc))
+		atomic_dec(&conn->cdc_pend_tx_wr);
+
+	return rc;
+}
+
+static int smcr_cdc_get_slot_and_msg_send(struct smc_connection *conn)
+{
+	struct smc_cdc_tx_pend *pend;
+	struct smc_wr_buf *wr_buf;
+	struct smc_link *link;
+	bool again = false;
+	int rc;
+
+again:
+	link = conn->lnk;
+	if (!smc_wr_tx_link_hold(link))
+		return -ENOLINK;
+	rc = smc_cdc_get_free_slot(conn, link, &wr_buf, NULL, &pend);
+	if (rc)
+		goto put_out;
+
+	spin_lock_bh(&conn->send_lock);
+	if (link != conn->lnk) {
+		/* link of connection changed, try again one time*/
+		spin_unlock_bh(&conn->send_lock);
+		smc_wr_tx_put_slot(link,
+				   (struct smc_wr_tx_pend_priv *)pend);
+		smc_wr_tx_link_put(link);
+		if (again)
+			return -ENOLINK;
+		again = true;
+		goto again;
+	}
+	rc = smc_cdc_msg_send(conn, wr_buf, pend);
+	spin_unlock_bh(&conn->send_lock);
+put_out:
+	smc_wr_tx_link_put(link);
+	return rc;
+}
+
+int smc_cdc_get_slot_and_msg_send(struct smc_connection *conn)
+{
+	int rc;
+
+	if (!smc_conn_lgr_valid(conn) ||
+	    (conn->lgr->is_smcd && conn->lgr->peer_shutdown))
+		return -EPIPE;
+
+	if (conn->lgr->is_smcd) {
+		spin_lock_bh(&conn->send_lock);
+		rc = smcd_cdc_msg_send(conn);
+		spin_unlock_bh(&conn->send_lock);
+	} else {
+		rc = smcr_cdc_get_slot_and_msg_send(conn);
+	}
+
+	return rc;
+}
+
+void smc_cdc_wait_pend_tx_wr(struct smc_connection *conn)
+{
+	wait_event(conn->cdc_pend_tx_wq, !atomic_read(&conn->cdc_pend_tx_wr));
+}
+
+/* Send a SMC-D CDC header.
+ * This increments the free space available in our send buffer.
+ * Also update the confirmed receive buffer with what was sent to the peer.
+ */
+int smcd_cdc_msg_send(struct smc_connection *conn)
+{
+	struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
+	union smc_host_cursor curs;
+	struct smcd_cdc_msg cdc;
+	int rc, diff;
+
+	memset(&cdc, 0, sizeof(cdc));
+	cdc.common.type = SMC_CDC_MSG_TYPE;
+	curs.acurs.counter = atomic64_read(&conn->local_tx_ctrl.prod.acurs);
+	cdc.prod.wrap = curs.wrap;
+	cdc.prod.count = curs.count;
+	curs.acurs.counter = atomic64_read(&conn->local_tx_ctrl.cons.acurs);
+	cdc.cons.wrap = curs.wrap;
+	cdc.cons.count = curs.count;
+	cdc.cons.prod_flags = conn->local_tx_ctrl.prod_flags;
+	cdc.cons.conn_state_flags = conn->local_tx_ctrl.conn_state_flags;
+	rc = smcd_tx_ism_write(conn, &cdc, sizeof(cdc), 0, 1);
+	if (rc)
+		return rc;
+	smc_curs_copy(&conn->rx_curs_confirmed, &curs, conn);
+	conn->local_rx_ctrl.prod_flags.cons_curs_upd_req = 0;
+
+	if (smc_ism_support_dmb_nocopy(conn->lgr->smcd))
+		/* if local sndbuf shares the same memory region with
+		 * peer DMB, then don't update the tx_curs_fin
+		 * and sndbuf_space until peer has consumed the data.
+		 */
+		return 0;
+
+	/* Calculate transmitted data and increment free send buffer space */
+	diff = smc_curs_diff(conn->sndbuf_desc->len, &conn->tx_curs_fin,
+			     &conn->tx_curs_sent);
+	/* increased by confirmed number of bytes */
+	smp_mb__before_atomic();
+	atomic_add(diff, &conn->sndbuf_space);
+	/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
+	smp_mb__after_atomic();
+	smc_curs_copy(&conn->tx_curs_fin, &conn->tx_curs_sent, conn);
+
+	smc_tx_sndbuf_nonfull(smc);
+	return 0;
+}
+
+/********************************* receive ***********************************/
+
+static inline bool smc_cdc_before(u16 seq1, u16 seq2)
+{
+	return (s16)(seq1 - seq2) < 0;
+}
+
+static void smc_cdc_handle_urg_data_arrival(struct smc_sock *smc,
+					    int *diff_prod)
+{
+	struct smc_connection *conn = &smc->conn;
+	char *base;
+
+	/* new data included urgent business */
+	smc_curs_copy(&conn->urg_curs, &conn->local_rx_ctrl.prod, conn);
+	conn->urg_state = SMC_URG_VALID;
+	if (!sock_flag(&smc->sk, SOCK_URGINLINE))
+		/* we'll skip the urgent byte, so don't account for it */
+		(*diff_prod)--;
+	base = (char *)conn->rmb_desc->cpu_addr + conn->rx_off;
+	if (conn->urg_curs.count)
+		conn->urg_rx_byte = *(base + conn->urg_curs.count - 1);
+	else
+		conn->urg_rx_byte = *(base + conn->rmb_desc->len - 1);
+	sk_send_sigurg(&smc->sk);
+}
+
+static void smc_cdc_msg_validate(struct smc_sock *smc, struct smc_cdc_msg *cdc,
+				 struct smc_link *link)
+{
+	struct smc_connection *conn = &smc->conn;
+	u16 recv_seq = ntohs(cdc->seqno);
+	s16 diff;
+
+	/* check that seqnum was seen before */
+	diff = conn->local_rx_ctrl.seqno - recv_seq;
+	if (diff < 0) { /* diff larger than 0x7fff */
+		/* drop connection */
+		conn->out_of_sync = 1;	/* prevent any further receives */
+		spin_lock_bh(&conn->send_lock);
+		conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
+		conn->lnk = link;
+		spin_unlock_bh(&conn->send_lock);
+		sock_hold(&smc->sk); /* sock_put in abort_work */
+		if (!queue_work(smc_close_wq, &conn->abort_work))
+			sock_put(&smc->sk);
+	}
+}
+
+static void smc_cdc_msg_recv_action(struct smc_sock *smc,
+				    struct smc_cdc_msg *cdc)
+{
+	union smc_host_cursor cons_old, prod_old;
+	struct smc_connection *conn = &smc->conn;
+	int diff_cons, diff_prod, diff_tx;
+
+	smc_curs_copy(&prod_old, &conn->local_rx_ctrl.prod, conn);
+	smc_curs_copy(&cons_old, &conn->local_rx_ctrl.cons, conn);
+	smc_cdc_msg_to_host(&conn->local_rx_ctrl, cdc, conn);
+
+	diff_cons = smc_curs_diff(conn->peer_rmbe_size, &cons_old,
+				  &conn->local_rx_ctrl.cons);
+	if (diff_cons) {
+		/* peer_rmbe_space is decreased during data transfer with RDMA
+		 * write
+		 */
+		smp_mb__before_atomic();
+		atomic_add(diff_cons, &conn->peer_rmbe_space);
+		/* guarantee 0 <= peer_rmbe_space <= peer_rmbe_size */
+		smp_mb__after_atomic();
+
+		/* if local sndbuf shares the same memory region with
+		 * peer RMB, then update tx_curs_fin and sndbuf_space
+		 * here since peer has already consumed the data.
+		 */
+		if (conn->lgr->is_smcd &&
+		    smc_ism_support_dmb_nocopy(conn->lgr->smcd)) {
+			/* Calculate consumed data and
+			 * increment free send buffer space.
+			 */
+			diff_tx = smc_curs_diff(conn->sndbuf_desc->len,
+						&conn->tx_curs_fin,
+						&conn->local_rx_ctrl.cons);
+			/* increase local sndbuf space and fin_curs */
+			smp_mb__before_atomic();
+			atomic_add(diff_tx, &conn->sndbuf_space);
+			/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
+			smp_mb__after_atomic();
+			smc_curs_copy(&conn->tx_curs_fin,
+				      &conn->local_rx_ctrl.cons, conn);
+
+			smc_tx_sndbuf_nonfull(smc);
+		}
+	}
+
+	diff_prod = smc_curs_diff(conn->rmb_desc->len, &prod_old,
+				  &conn->local_rx_ctrl.prod);
+	if (diff_prod) {
+		if (conn->local_rx_ctrl.prod_flags.urg_data_present)
+			smc_cdc_handle_urg_data_arrival(smc, &diff_prod);
+		/* bytes_to_rcv is decreased in smc_recvmsg */
+		smp_mb__before_atomic();
+		atomic_add(diff_prod, &conn->bytes_to_rcv);
+		/* guarantee 0 <= bytes_to_rcv <= rmb_desc->len */
+		smp_mb__after_atomic();
+		smc->sk.sk_data_ready(&smc->sk);
+	} else {
+		if (conn->local_rx_ctrl.prod_flags.write_blocked)
+			smc->sk.sk_data_ready(&smc->sk);
+		if (conn->local_rx_ctrl.prod_flags.urg_data_pending)
+			conn->urg_state = SMC_URG_NOTYET;
+	}
+
+	/* trigger sndbuf consumer: RDMA write into peer RMBE and CDC */
+	if ((diff_cons && smc_tx_prepared_sends(conn)) ||
+	    conn->local_rx_ctrl.prod_flags.cons_curs_upd_req ||
+	    conn->local_rx_ctrl.prod_flags.urg_data_pending) {
+		if (!sock_owned_by_user(&smc->sk))
+			smc_tx_pending(conn);
+		else
+			conn->tx_in_release_sock = true;
+	}
+
+	if (diff_cons && conn->urg_tx_pend &&
+	    atomic_read(&conn->peer_rmbe_space) == conn->peer_rmbe_size) {
+		/* urg data confirmed by peer, indicate we're ready for more */
+		conn->urg_tx_pend = false;
+		smc->sk.sk_write_space(&smc->sk);
+	}
+
+	if (conn->local_rx_ctrl.conn_state_flags.peer_conn_abort) {
+		smc->sk.sk_err = ECONNRESET;
+		conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
+	}
+	if (smc_cdc_rxed_any_close_or_senddone(conn)) {
+		smc->sk.sk_shutdown |= RCV_SHUTDOWN;
+		if (smc->clcsock && smc->clcsock->sk)
+			smc->clcsock->sk->sk_shutdown |= RCV_SHUTDOWN;
+		smc_sock_set_flag(&smc->sk, SOCK_DONE);
+		sock_hold(&smc->sk); /* sock_put in close_work */
+		if (!queue_work(smc_close_wq, &conn->close_work))
+			sock_put(&smc->sk);
+	}
+}
+
+/* called under tasklet context */
+static void smc_cdc_msg_recv(struct smc_sock *smc, struct smc_cdc_msg *cdc)
+{
+	sock_hold(&smc->sk);
+	bh_lock_sock(&smc->sk);
+	smc_cdc_msg_recv_action(smc, cdc);
+	bh_unlock_sock(&smc->sk);
+	sock_put(&smc->sk); /* no free sk in softirq-context */
+}
+
+/* Schedule a tasklet for this connection. Triggered from the ISM device IRQ
+ * handler to indicate update in the DMBE.
+ *
+ * Context:
+ * - tasklet context
+ */
+static void smcd_cdc_rx_tsklet(struct tasklet_struct *t)
+{
+	struct smc_connection *conn = from_tasklet(conn, t, rx_tsklet);
+	struct smcd_cdc_msg *data_cdc;
+	struct smcd_cdc_msg cdc;
+	struct smc_sock *smc;
+
+	if (!conn || conn->killed)
+		return;
+
+	data_cdc = (struct smcd_cdc_msg *)conn->rmb_desc->cpu_addr;
+	smcd_curs_copy(&cdc.prod, &data_cdc->prod, conn);
+	smcd_curs_copy(&cdc.cons, &data_cdc->cons, conn);
+	smc = container_of(conn, struct smc_sock, conn);
+	smc_cdc_msg_recv(smc, (struct smc_cdc_msg *)&cdc);
+}
+
+/* Initialize receive tasklet. Called from ISM device IRQ handler to start
+ * receiver side.
+ */
+void smcd_cdc_rx_init(struct smc_connection *conn)
+{
+	tasklet_setup(&conn->rx_tsklet, smcd_cdc_rx_tsklet);
+}
+
+/***************************** init, exit, misc ******************************/
+
+static void smc_cdc_rx_handler(struct ib_wc *wc, void *buf)
+{
+	struct smc_link *link = (struct smc_link *)wc->qp->qp_context;
+	struct smc_cdc_msg *cdc = buf;
+	struct smc_sock *smc = NULL;
+	struct smc_connection *conn;
+	struct smc_link_group *lgr;
+
+	if (wc->byte_len < offsetof(struct smc_cdc_msg, reserved))
+		return; /* short message */
+	if (cdc->len != SMC_WR_TX_SIZE)
+		return; /* invalid message */
+
+	/* lookup connection */
+	lgr = smc_get_lgr(link);
+	read_lock_bh(&lgr->conns_lock);
+	conn = smc_lgr_find_conn(ntohl(cdc->token), lgr);
+	if (!conn || conn->out_of_sync) {
+		read_unlock_bh(&lgr->conns_lock);
+		return;
+	}
+	smc = container_of(conn, struct smc_sock, conn);
+	sock_hold(&smc->sk);
+	read_unlock_bh(&lgr->conns_lock);
+
+	if (cdc->prod_flags.failover_validation) {
+		smc_cdc_msg_validate(smc, cdc, link);
+		goto out;
+	}
+	if (smc_cdc_before(ntohs(cdc->seqno),
+			   conn->local_rx_ctrl.seqno))
+		/* received seqno is old */
+		goto out;
+
+	smc_cdc_msg_recv(smc, cdc);
+out:
+	sock_put(&smc->sk);
+}
+
+static struct smc_wr_rx_handler smc_cdc_rx_handlers[] = {
+	{
+		.handler	= smc_cdc_rx_handler,
+		.type		= SMC_CDC_MSG_TYPE
+	},
+	{
+		.handler	= NULL,
+	}
+};
+
+int __init smc_cdc_init(void)
+{
+	struct smc_wr_rx_handler *handler;
+	int rc = 0;
+
+	for (handler = smc_cdc_rx_handlers; handler->handler; handler++) {
+		INIT_HLIST_NODE(&handler->list);
+		rc = smc_wr_rx_register_handler(handler);
+		if (rc)
+			break;
+	}
+	return rc;
+}

linux-7.2.3/net/smc/smc_cdc.h

@@ -0,0 +1,305 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Connection Data Control (CDC)
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef SMC_CDC_H
+#define SMC_CDC_H
+
+#include <linux/kernel.h> /* max_t */
+#include <linux/atomic.h>
+#include <linux/in.h>
+#include <linux/compiler.h>
+
+#include "smc.h"
+#include "smc_core.h"
+#include "smc_wr.h"
+
+#define	SMC_CDC_MSG_TYPE		0xFE
+
+/* in network byte order */
+union smc_cdc_cursor {		/* SMC cursor */
+	struct {
+		__be16	reserved;
+		__be16	wrap;
+		__be32	count;
+	};
+#ifdef KERNEL_HAS_ATOMIC64
+	atomic64_t	acurs;		/* for atomic processing */
+#else
+	u64		acurs;		/* for atomic processing */
+#endif
+} __aligned(8);
+
+/* in network byte order */
+struct smc_cdc_msg {
+	struct smc_wr_rx_hdr		common; /* .type = 0xFE */
+	u8				len;	/* 44 */
+	__be16				seqno;
+	__be32				token;
+	union smc_cdc_cursor		prod;
+	union smc_cdc_cursor		cons;	/* piggy backed "ack" */
+	struct smc_cdc_producer_flags	prod_flags;
+	struct smc_cdc_conn_state_flags	conn_state_flags;
+	u8				reserved[18];
+};
+
+/* SMC-D cursor format */
+union smcd_cdc_cursor {
+	struct {
+		u16	wrap;
+		u32	count;
+		struct smc_cdc_producer_flags	prod_flags;
+		struct smc_cdc_conn_state_flags	conn_state_flags;
+	} __packed;
+#ifdef KERNEL_HAS_ATOMIC64
+	atomic64_t		acurs;		/* for atomic processing */
+#else
+	u64			acurs;		/* for atomic processing */
+#endif
+} __aligned(8);
+
+/* CDC message for SMC-D */
+struct smcd_cdc_msg {
+	struct smc_wr_rx_hdr common;	/* Type = 0xFE */
+	u8 res1[7];
+	union smcd_cdc_cursor	prod;
+	union smcd_cdc_cursor	cons;
+	u8 res3[8];
+} __aligned(8);
+
+static inline bool smc_cdc_rxed_any_close(struct smc_connection *conn)
+{
+	return conn->local_rx_ctrl.conn_state_flags.peer_conn_abort ||
+	       conn->local_rx_ctrl.conn_state_flags.peer_conn_closed;
+}
+
+static inline bool smc_cdc_rxed_any_close_or_senddone(
+	struct smc_connection *conn)
+{
+	return smc_cdc_rxed_any_close(conn) ||
+	       conn->local_rx_ctrl.conn_state_flags.peer_done_writing;
+}
+
+static inline void smc_curs_add(int size, union smc_host_cursor *curs,
+				int value)
+{
+	curs->count += value;
+	if (curs->count >= size) {
+		curs->wrap++;
+		curs->count -= size;
+	}
+}
+
+/* Copy cursor src into tgt */
+static inline void smc_curs_copy(union smc_host_cursor *tgt,
+				 union smc_host_cursor *src,
+				 struct smc_connection *conn)
+{
+#ifndef KERNEL_HAS_ATOMIC64
+	unsigned long flags;
+
+	spin_lock_irqsave(&conn->acurs_lock, flags);
+	tgt->acurs = src->acurs;
+	spin_unlock_irqrestore(&conn->acurs_lock, flags);
+#else
+	atomic64_set(&tgt->acurs, atomic64_read(&src->acurs));
+#endif
+}
+
+static inline void smc_curs_copy_net(union smc_cdc_cursor *tgt,
+				     union smc_cdc_cursor *src,
+				     struct smc_connection *conn)
+{
+#ifndef KERNEL_HAS_ATOMIC64
+	unsigned long flags;
+
+	spin_lock_irqsave(&conn->acurs_lock, flags);
+	tgt->acurs = src->acurs;
+	spin_unlock_irqrestore(&conn->acurs_lock, flags);
+#else
+	atomic64_set(&tgt->acurs, atomic64_read(&src->acurs));
+#endif
+}
+
+static inline void smcd_curs_copy(union smcd_cdc_cursor *tgt,
+				  union smcd_cdc_cursor *src,
+				  struct smc_connection *conn)
+{
+#ifndef KERNEL_HAS_ATOMIC64
+	unsigned long flags;
+
+	spin_lock_irqsave(&conn->acurs_lock, flags);
+	tgt->acurs = src->acurs;
+	spin_unlock_irqrestore(&conn->acurs_lock, flags);
+#else
+	atomic64_set(&tgt->acurs, atomic64_read(&src->acurs));
+#endif
+}
+
+/* calculate cursor difference between old and new, where old <= new and
+ * difference cannot exceed size
+ */
+static inline int smc_curs_diff(unsigned int size,
+				union smc_host_cursor *old,
+				union smc_host_cursor *new)
+{
+	if (old->wrap != new->wrap)
+		return max_t(int, 0,
+			     ((size - old->count) + new->count));
+
+	return max_t(int, 0, (new->count - old->count));
+}
+
+/* calculate cursor difference between old and new - returns negative
+ * value in case old > new
+ */
+static inline int smc_curs_comp(unsigned int size,
+				union smc_host_cursor *old,
+				union smc_host_cursor *new)
+{
+	if (old->wrap > new->wrap ||
+	    (old->wrap == new->wrap && old->count > new->count))
+		return -smc_curs_diff(size, new, old);
+	return smc_curs_diff(size, old, new);
+}
+
+/* calculate cursor difference between old and new, where old <= new and
+ * difference may exceed size
+ */
+static inline int smc_curs_diff_large(unsigned int size,
+				      union smc_host_cursor *old,
+				      union smc_host_cursor *new)
+{
+	if (old->wrap < new->wrap)
+		return min_t(int,
+			     (size - old->count) + new->count +
+			     (new->wrap - old->wrap - 1) * size,
+			     size);
+
+	if (old->wrap > new->wrap) /* wrap has switched from 0xffff to 0x0000 */
+		return min_t(int,
+			     (size - old->count) + new->count +
+			     (new->wrap + 0xffff - old->wrap) * size,
+			     size);
+
+	return max_t(int, 0, (new->count - old->count));
+}
+
+static inline void smc_host_cursor_to_cdc(union smc_cdc_cursor *peer,
+					  union smc_host_cursor *local,
+					  union smc_host_cursor *save,
+					  struct smc_connection *conn)
+{
+	smc_curs_copy(save, local, conn);
+	peer->count = htonl(save->count);
+	peer->wrap = htons(save->wrap);
+	/* peer->reserved = htons(0); must be ensured by caller */
+}
+
+static inline void smc_host_msg_to_cdc(struct smc_cdc_msg *peer,
+				       struct smc_connection *conn,
+				       union smc_host_cursor *save)
+{
+	struct smc_host_cdc_msg *local = &conn->local_tx_ctrl;
+
+	peer->common.type = local->common.type;
+	peer->len = local->len;
+	peer->seqno = htons(local->seqno);
+	peer->token = htonl(local->token);
+	smc_host_cursor_to_cdc(&peer->prod, &local->prod, save, conn);
+	smc_host_cursor_to_cdc(&peer->cons, &local->cons, save, conn);
+	peer->prod_flags = local->prod_flags;
+	peer->conn_state_flags = local->conn_state_flags;
+}
+
+static inline void smc_cdc_cursor_to_host(union smc_host_cursor *local,
+					  union smc_cdc_cursor *peer,
+					  struct smc_connection *conn)
+{
+	union smc_host_cursor temp, old;
+	union smc_cdc_cursor net;
+
+	smc_curs_copy(&old, local, conn);
+	smc_curs_copy_net(&net, peer, conn);
+	temp.count = ntohl(net.count);
+	temp.wrap = ntohs(net.wrap);
+	if ((old.wrap > temp.wrap) && temp.wrap)
+		return;
+	if ((old.wrap == temp.wrap) &&
+	    (old.count > temp.count))
+		return;
+	smc_curs_copy(local, &temp, conn);
+}
+
+static inline void smcr_cdc_msg_to_host(struct smc_host_cdc_msg *local,
+					struct smc_cdc_msg *peer,
+					struct smc_connection *conn)
+{
+	local->common.type = peer->common.type;
+	local->len = peer->len;
+	local->seqno = ntohs(peer->seqno);
+	local->token = ntohl(peer->token);
+	smc_cdc_cursor_to_host(&local->prod, &peer->prod, conn);
+	smc_cdc_cursor_to_host(&local->cons, &peer->cons, conn);
+	local->prod_flags = peer->prod_flags;
+	local->conn_state_flags = peer->conn_state_flags;
+}
+
+static inline void smcd_cdc_msg_to_host(struct smc_host_cdc_msg *local,
+					struct smcd_cdc_msg *peer,
+					struct smc_connection *conn)
+{
+	union smc_host_cursor temp;
+
+	temp.wrap = peer->prod.wrap;
+	temp.count = peer->prod.count;
+	smc_curs_copy(&local->prod, &temp, conn);
+
+	temp.wrap = peer->cons.wrap;
+	temp.count = peer->cons.count;
+	smc_curs_copy(&local->cons, &temp, conn);
+	local->prod_flags = peer->cons.prod_flags;
+	local->conn_state_flags = peer->cons.conn_state_flags;
+}
+
+static inline void smc_cdc_msg_to_host(struct smc_host_cdc_msg *local,
+				       struct smc_cdc_msg *peer,
+				       struct smc_connection *conn)
+{
+	if (conn->lgr->is_smcd)
+		smcd_cdc_msg_to_host(local, (struct smcd_cdc_msg *)peer, conn);
+	else
+		smcr_cdc_msg_to_host(local, peer, conn);
+}
+
+struct smc_cdc_tx_pend {
+	struct smc_connection	*conn;		/* socket connection */
+	union smc_host_cursor	cursor;		/* tx sndbuf cursor sent */
+	union smc_host_cursor	p_cursor;	/* rx RMBE cursor produced */
+	u16			ctrl_seq;	/* conn. tx sequence # */
+};
+
+int smc_cdc_get_free_slot(struct smc_connection *conn,
+			  struct smc_link *link,
+			  struct smc_wr_buf **wr_buf,
+			  struct smc_rdma_wr **wr_rdma_buf,
+			  struct smc_cdc_tx_pend **pend);
+void smc_cdc_wait_pend_tx_wr(struct smc_connection *conn);
+int smc_cdc_msg_send(struct smc_connection *conn, struct smc_wr_buf *wr_buf,
+		     struct smc_cdc_tx_pend *pend);
+int smc_cdc_get_slot_and_msg_send(struct smc_connection *conn);
+int smcd_cdc_msg_send(struct smc_connection *conn);
+int smcr_cdc_msg_send_validation(struct smc_connection *conn,
+				 struct smc_cdc_tx_pend *pend,
+				 struct smc_wr_buf *wr_buf);
+int smc_cdc_init(void) __init;
+void smcd_cdc_rx_init(struct smc_connection *conn);
+
+#endif /* SMC_CDC_H */

linux-7.2.3/net/smc/smc_clc.h

@@ -0,0 +1,477 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  CLC (connection layer control) handshake over initial TCP socket to
+ *  prepare for RDMA traffic
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef _SMC_CLC_H
+#define _SMC_CLC_H
+
+#include <rdma/ib_verbs.h>
+#include <linux/smc.h>
+
+#include "smc.h"
+#include "smc_netlink.h"
+
+#define SMC_CLC_PROPOSAL	0x01
+#define SMC_CLC_ACCEPT		0x02
+#define SMC_CLC_CONFIRM		0x03
+#define SMC_CLC_DECLINE		0x04
+
+#define SMC_TYPE_R		0		/* SMC-R only		      */
+#define SMC_TYPE_D		1		/* SMC-D only		      */
+#define SMC_TYPE_N		2		/* neither SMC-R nor SMC-D    */
+#define SMC_TYPE_B		3		/* SMC-R and SMC-D	      */
+#define CLC_WAIT_TIME		(6 * HZ)	/* max. wait time on clcsock  */
+#define CLC_WAIT_TIME_SHORT	HZ		/* short wait time on clcsock */
+#define SMC_CLC_DECL_MEM	0x01010000  /* insufficient memory resources  */
+#define SMC_CLC_DECL_TIMEOUT_CL	0x02010000  /* timeout w4 QP confirm link     */
+#define SMC_CLC_DECL_TIMEOUT_AL	0x02020000  /* timeout w4 QP add link	      */
+#define SMC_CLC_DECL_CNFERR	0x03000000  /* configuration error            */
+#define SMC_CLC_DECL_PEERNOSMC	0x03010000  /* peer did not indicate SMC      */
+#define SMC_CLC_DECL_IPSEC	0x03020000  /* IPsec usage		      */
+#define SMC_CLC_DECL_NOSMCDEV	0x03030000  /* no SMC device found (R or D)   */
+#define SMC_CLC_DECL_NOSMCDDEV	0x03030001  /* no SMC-D device found	      */
+#define SMC_CLC_DECL_NOSMCRDEV	0x03030002  /* no SMC-R device found	      */
+#define SMC_CLC_DECL_NOISM2SUPP	0x03030003  /* hardware has no ISMv2 support  */
+#define SMC_CLC_DECL_NOV2EXT	0x03030004  /* peer sent no clc v2 extension  */
+#define SMC_CLC_DECL_NOV2DEXT	0x03030005  /* peer sent no clc SMC-Dv2 ext.  */
+#define SMC_CLC_DECL_NOSEID	0x03030006  /* peer sent no SEID	      */
+#define SMC_CLC_DECL_NOSMCD2DEV	0x03030007  /* no SMC-Dv2 device found	      */
+#define SMC_CLC_DECL_NOUEID	0x03030008  /* peer sent no UEID	      */
+#define SMC_CLC_DECL_RELEASEERR	0x03030009  /* release version negotiate failed */
+#define SMC_CLC_DECL_MAXCONNERR	0x0303000a  /* max connections negotiate failed */
+#define SMC_CLC_DECL_MAXLINKERR	0x0303000b  /* max links negotiate failed */
+#define SMC_CLC_DECL_MODEUNSUPP	0x03040000  /* smc modes do not match (R or D)*/
+#define SMC_CLC_DECL_RMBE_EC	0x03050000  /* peer has eyecatcher in RMBE    */
+#define SMC_CLC_DECL_OPTUNSUPP	0x03060000  /* fastopen sockopt not supported */
+#define SMC_CLC_DECL_DIFFPREFIX	0x03070000  /* IP prefix / subnet mismatch    */
+#define SMC_CLC_DECL_GETVLANERR	0x03080000  /* err to get vlan id of ip device*/
+#define SMC_CLC_DECL_ISMVLANERR	0x03090000  /* err to reg vlan id on ism dev  */
+#define SMC_CLC_DECL_NOACTLINK	0x030a0000  /* no active smc-r link in lgr    */
+#define SMC_CLC_DECL_NOSRVLINK	0x030b0000  /* SMC-R link from srv not found  */
+#define SMC_CLC_DECL_VERSMISMAT	0x030c0000  /* SMC version mismatch	      */
+#define SMC_CLC_DECL_MAX_DMB	0x030d0000  /* SMC-D DMB limit exceeded       */
+#define SMC_CLC_DECL_NOROUTE	0x030e0000  /* SMC-Rv2 conn. no route to peer */
+#define SMC_CLC_DECL_NOINDIRECT	0x030f0000  /* SMC-Rv2 conn. indirect mismatch*/
+#define SMC_CLC_DECL_SYNCERR	0x04000000  /* synchronization error          */
+#define SMC_CLC_DECL_PEERDECL	0x05000000  /* peer declined during handshake */
+#define SMC_CLC_DECL_INTERR	0x09990000  /* internal error		      */
+#define SMC_CLC_DECL_ERR_RTOK	0x09990001  /*	 rtoken handling failed       */
+#define SMC_CLC_DECL_ERR_RDYLNK	0x09990002  /*	 ib ready link failed	      */
+#define SMC_CLC_DECL_ERR_REGBUF	0x09990003  /*	 reg rdma bufs failed	      */
+
+#define SMC_FIRST_CONTACT_MASK	0b10	/* first contact bit within typev2 */
+
+struct smc_clc_msg_hdr {	/* header1 of clc messages */
+	u8 eyecatcher[4];	/* eye catcher */
+	u8 type;		/* proposal / accept / confirm / decline */
+	__be16 length;
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8 version : 4,
+	   typev2  : 2,
+	   typev1  : 2;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8 typev1  : 2,
+	   typev2  : 2,
+	   version : 4;
+#endif
+} __packed;			/* format defined in RFC7609 */
+
+struct smc_clc_msg_trail {	/* trailer of clc messages */
+	u8 eyecatcher[4];
+};
+
+struct smc_clc_msg_local {	/* header2 of clc messages */
+	u8 id_for_peer[SMC_SYSTEMID_LEN]; /* unique system id */
+	u8 gid[16];		/* gid of ib_device port */
+	u8 mac[6];		/* mac of ib_device port */
+};
+
+/* Struct would be 4 byte aligned, but it is used in an array that is sent
+ * to peers and must conform to RFC7609, hence we need to use packed here.
+ */
+struct smc_clc_ipv6_prefix {
+	struct in6_addr prefix;
+	u8 prefix_len;
+} __packed;			/* format defined in RFC7609 */
+
+#if defined(__BIG_ENDIAN_BITFIELD)
+struct smc_clc_v2_flag {
+	u8 release : 4,
+	   rsvd    : 3,
+	   seid    : 1;
+};
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+struct smc_clc_v2_flag {
+	u8 seid   : 1,
+	rsvd      : 3,
+	release   : 4;
+};
+#endif
+
+struct smc_clnt_opts_area_hdr {
+	u8 eid_cnt;		/* number of user defined EIDs */
+	u8 ism_gid_cnt;		/* number of ISMv2 GIDs */
+	u8 reserved1;
+	struct smc_clc_v2_flag flag;
+	u8 reserved2[2];
+	__be16 smcd_v2_ext_offset; /* SMC-Dv2 Extension Offset */
+};
+
+struct smc_clc_smcd_gid_chid {
+	__be64 gid;		/* ISM GID */
+	__be16 chid;		/* ISMv2 CHID */
+} __packed;		/* format defined in
+			 * IBM Shared Memory Communications Version 2
+			 * (https://www.ibm.com/support/pages/node/6326337)
+			 */
+
+struct smc_clc_v2_extension {
+	/* New members must be added within the struct_group() macro below. */
+	struct_group_tagged(smc_clc_v2_extension_fixed, fixed,
+		struct smc_clnt_opts_area_hdr hdr;
+		u8 roce[16];		/* RoCEv2 GID */
+		u8 max_conns;
+		u8 max_links;
+		__be16 feature_mask;
+		u8 reserved[12];
+	);
+	u8 user_eids[][SMC_MAX_EID_LEN];
+};
+static_assert(offsetof(struct smc_clc_v2_extension, user_eids) == sizeof(struct smc_clc_v2_extension_fixed),
+	      "struct member likely outside of struct_group_tagged()");
+
+struct smc_clc_msg_proposal_prefix {	/* prefix part of clc proposal message*/
+	__be32 outgoing_subnet;	/* subnet mask */
+	u8 prefix_len;		/* number of significant bits in mask */
+	u8 reserved[2];
+	u8 ipv6_prefixes_cnt;	/* number of IPv6 prefixes in prefix array */
+} __aligned(4);
+
+struct smc_clc_msg_smcd {	/* SMC-D GID information */
+	struct smc_clc_smcd_gid_chid ism; /* ISM native GID+CHID of requester */
+	__be16 v2_ext_offset;	/* SMC Version 2 Extension Offset */
+	u8 vendor_oui[3];	/* vendor organizationally unique identifier */
+	u8 vendor_exp_options[5];
+	u8 reserved[20];
+};
+
+struct smc_clc_smcd_v2_extension {
+	/* New members must be added within the struct_group() macro below. */
+	struct_group_tagged(smc_clc_smcd_v2_extension_fixed, fixed,
+		u8 system_eid[SMC_MAX_EID_LEN];
+		u8 reserved[16];
+	);
+	struct smc_clc_smcd_gid_chid gidchid[];
+};
+static_assert(offsetof(struct smc_clc_smcd_v2_extension, gidchid) == sizeof(struct smc_clc_smcd_v2_extension_fixed),
+	      "struct member likely outside of struct_group_tagged()");
+
+struct smc_clc_msg_proposal {	/* clc proposal message sent by Linux */
+	struct smc_clc_msg_hdr hdr;
+	struct smc_clc_msg_local lcl;
+	__be16 iparea_offset;	/* offset to IP address information area */
+} __aligned(4);
+
+#define SMC_CLC_MAX_V6_PREFIX		8
+#define SMC_CLC_MAX_UEID		8
+#define SMCD_CLC_MAX_V2_GID_ENTRIES	8 /* max # of CHID-GID entries in CLC
+					   * proposal SMC-Dv2 extension.
+					   * each ISM device takes one entry and
+					   * each Emulated-ISM takes two entries
+					   */
+
+struct smc_clc_msg_proposal_area {
+	struct smc_clc_msg_proposal		pclc_base;
+	struct smc_clc_msg_smcd			pclc_smcd;
+	struct smc_clc_msg_proposal_prefix	pclc_prfx;
+	struct smc_clc_ipv6_prefix	pclc_prfx_ipv6[SMC_CLC_MAX_V6_PREFIX];
+	struct smc_clc_v2_extension_fixed	pclc_v2_ext;
+	u8			user_eids[SMC_CLC_MAX_UEID][SMC_MAX_EID_LEN];
+	struct smc_clc_smcd_v2_extension_fixed	pclc_smcd_v2_ext;
+	struct smc_clc_smcd_gid_chid
+				pclc_gidchids[SMCD_CLC_MAX_V2_GID_ENTRIES];
+	struct smc_clc_msg_trail		pclc_trl;
+};
+
+struct smcr_clc_msg_accept_confirm {	/* SMCR accept/confirm */
+	struct smc_clc_msg_local lcl;
+	u8 qpn[3];			/* QP number */
+	__be32 rmb_rkey;		/* RMB rkey */
+	u8 rmbe_idx;			/* Index of RMBE in RMB */
+	__be32 rmbe_alert_token;	/* unique connection id */
+ #if defined(__BIG_ENDIAN_BITFIELD)
+	u8 rmbe_size : 4,		/* buf size (compressed) */
+	   qp_mtu   : 4;		/* QP mtu */
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8 qp_mtu   : 4,
+	   rmbe_size : 4;
+#endif
+	u8 reserved;
+	__be64 rmb_dma_addr;	/* RMB virtual address */
+	u8 reserved2;
+	u8 psn[3];		/* packet sequence number */
+} __packed;
+
+struct smcd_clc_msg_accept_confirm_common {	/* SMCD accept/confirm */
+	__be64 gid;		/* Sender GID */
+	__be64 token;		/* DMB token */
+	u8 dmbe_idx;		/* DMBE index */
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8 dmbe_size : 4,	/* buf size (compressed) */
+	   reserved3 : 4;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8 reserved3 : 4,
+	   dmbe_size : 4;
+#endif
+	u16 reserved4;
+	__be32 linkid;		/* Link identifier */
+} __packed;
+
+#define SMC_CLC_OS_ZOS		1
+#define SMC_CLC_OS_LINUX	2
+#define SMC_CLC_OS_AIX		3
+
+struct smc_clc_first_contact_ext {
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8 v2_direct : 1,
+	   reserved  : 7;
+	u8 os_type : 4,
+	   release : 4;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8 reserved  : 7,
+	   v2_direct : 1;
+	u8 release : 4,
+	   os_type : 4;
+#endif
+	u8 reserved2[2];
+	u8 hostname[SMC_MAX_HOSTNAME_LEN];
+};
+
+struct smc_clc_first_contact_ext_v2x {
+	struct smc_clc_first_contact_ext fce_v2_base;
+	union {
+		struct {
+			u8 max_conns; /* for SMC-R only */
+			u8 max_links; /* for SMC-R only */
+		};
+		u8 reserved3[2];	/* for SMC-D only */
+	};
+	__be16 feature_mask;
+	__be32 vendor_exp_options;
+	u8 reserved4[8];
+} __packed;		/* format defined in
+			 * IBM Shared Memory Communications Version 2 (Third Edition)
+			 * (https://www.ibm.com/support/pages/node/7009315)
+			 */
+
+struct smc_clc_fce_gid_ext {
+	u8 gid_cnt;
+	u8 reserved2[3];
+	u8 gid[][SMC_GID_SIZE];
+};
+
+struct smc_clc_msg_accept_confirm {	/* clc accept / confirm message */
+	struct smc_clc_msg_hdr hdr;
+	union {
+		struct { /* SMC-R */
+			struct smcr_clc_msg_accept_confirm r0;
+			struct { /* v2 only */
+				u8 eid[SMC_MAX_EID_LEN];
+				u8 reserved6[8];
+			} __packed r1;
+		};
+		struct { /* SMC-D */
+			struct smcd_clc_msg_accept_confirm_common d0;
+			struct { /* v2 only, but 12 bytes reserved in v1 */
+				__be16 chid;
+				u8 eid[SMC_MAX_EID_LEN];
+				__be64 gid_ext;
+			} __packed d1;
+		};
+	};
+};
+
+struct smc_clc_msg_decline {	/* clc decline message */
+	struct smc_clc_msg_hdr hdr;
+	u8 id_for_peer[SMC_SYSTEMID_LEN]; /* sender peer_id */
+	__be32 peer_diagnosis;	/* diagnosis information */
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8 os_type  : 4,
+	   reserved : 4;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8 reserved : 4,
+	   os_type  : 4;
+#endif
+	u8 reserved2[3];
+	struct smc_clc_msg_trail trl; /* eye catcher "SMCD" or "SMCR" EBCDIC */
+} __aligned(4);
+
+#define SMC_DECL_DIAG_COUNT_V2	4 /* no. of additional peer diagnosis codes */
+
+struct smc_clc_msg_decline_v2 {	/* clc decline message */
+	struct smc_clc_msg_hdr hdr;
+	u8 id_for_peer[SMC_SYSTEMID_LEN]; /* sender peer_id */
+	__be32 peer_diagnosis;	/* diagnosis information */
+#if defined(__BIG_ENDIAN_BITFIELD)
+	u8 os_type  : 4,
+	   reserved : 4;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	u8 reserved : 4,
+	   os_type  : 4;
+#endif
+	u8 reserved2[3];
+	__be32 peer_diagnosis_v2[SMC_DECL_DIAG_COUNT_V2];
+	struct smc_clc_msg_trail trl; /* eye catcher "SMCD" or "SMCR" EBCDIC */
+} __aligned(4);
+
+/* determine start of the prefix area within the proposal message */
+static inline struct smc_clc_msg_proposal_prefix *
+smc_clc_proposal_get_prefix(struct smc_clc_msg_proposal *pclc)
+{
+	u16 offset = ntohs(pclc->iparea_offset);
+
+	if (offset > sizeof(struct smc_clc_msg_smcd))
+		return NULL;
+	return (struct smc_clc_msg_proposal_prefix *)
+	       ((u8 *)pclc + sizeof(*pclc) + offset);
+}
+
+static inline bool smcr_indicated(int smc_type)
+{
+	return smc_type == SMC_TYPE_R || smc_type == SMC_TYPE_B;
+}
+
+static inline bool smcd_indicated(int smc_type)
+{
+	return smc_type == SMC_TYPE_D || smc_type == SMC_TYPE_B;
+}
+
+static inline u8 smc_indicated_type(int is_smcd, int is_smcr)
+{
+	if (is_smcd && is_smcr)
+		return SMC_TYPE_B;
+	if (is_smcd)
+		return SMC_TYPE_D;
+	if (is_smcr)
+		return SMC_TYPE_R;
+	return SMC_TYPE_N;
+}
+
+/* get SMC-D info from proposal message */
+static inline struct smc_clc_msg_smcd *
+smc_get_clc_msg_smcd(struct smc_clc_msg_proposal *prop)
+{
+	if (smcd_indicated(prop->hdr.typev1) &&
+	    ntohs(prop->iparea_offset) != sizeof(struct smc_clc_msg_smcd))
+		return NULL;
+
+	return (struct smc_clc_msg_smcd *)(prop + 1);
+}
+
+static inline struct smc_clc_v2_extension *
+smc_get_clc_v2_ext(struct smc_clc_msg_proposal *prop)
+{
+	struct smc_clc_msg_smcd *prop_smcd = smc_get_clc_msg_smcd(prop);
+	u16 max_offset;
+
+	max_offset = offsetof(struct smc_clc_msg_proposal_area, pclc_v2_ext) -
+		     offsetof(struct smc_clc_msg_proposal_area, pclc_smcd) -
+		     offsetofend(struct smc_clc_msg_smcd, v2_ext_offset);
+
+	if (!prop_smcd || !ntohs(prop_smcd->v2_ext_offset) ||
+	    ntohs(prop_smcd->v2_ext_offset) > max_offset)
+		return NULL;
+
+	return (struct smc_clc_v2_extension *)
+	       ((u8 *)prop_smcd +
+	       offsetof(struct smc_clc_msg_smcd, v2_ext_offset) +
+	       sizeof(prop_smcd->v2_ext_offset) +
+	       ntohs(prop_smcd->v2_ext_offset));
+}
+
+static inline struct smc_clc_smcd_v2_extension *
+smc_get_clc_smcd_v2_ext(struct smc_clc_v2_extension *prop_v2ext)
+{
+	u16 max_offset = offsetof(struct smc_clc_msg_proposal_area, pclc_smcd_v2_ext) -
+		offsetof(struct smc_clc_msg_proposal_area, pclc_v2_ext) -
+		offsetof(struct smc_clc_v2_extension, hdr) -
+		offsetofend(struct smc_clnt_opts_area_hdr, smcd_v2_ext_offset);
+
+	if (!prop_v2ext)
+		return NULL;
+	if (!ntohs(prop_v2ext->hdr.smcd_v2_ext_offset) ||
+	    ntohs(prop_v2ext->hdr.smcd_v2_ext_offset) > max_offset)
+		return NULL;
+
+	return (struct smc_clc_smcd_v2_extension *)
+		((u8 *)prop_v2ext +
+		 offsetof(struct smc_clc_v2_extension, hdr) +
+		 offsetof(struct smc_clnt_opts_area_hdr, smcd_v2_ext_offset) +
+		 sizeof(prop_v2ext->hdr.smcd_v2_ext_offset) +
+		 ntohs(prop_v2ext->hdr.smcd_v2_ext_offset));
+}
+
+static inline struct smc_clc_first_contact_ext *
+smc_get_clc_first_contact_ext(struct smc_clc_msg_accept_confirm *clc,
+			      bool is_smcd)
+{
+	int clc_v2_len;
+
+	if (clc->hdr.version == SMC_V1 ||
+	    !(clc->hdr.typev2 & SMC_FIRST_CONTACT_MASK))
+		return NULL;
+
+	if (is_smcd)
+		clc_v2_len =
+			offsetofend(struct smc_clc_msg_accept_confirm, d1);
+	else
+		clc_v2_len =
+			offsetofend(struct smc_clc_msg_accept_confirm, r1);
+
+	return (struct smc_clc_first_contact_ext *)(((u8 *)clc) + clc_v2_len);
+}
+
+struct smcd_dev;
+struct smc_init_info;
+
+int smc_clc_prfx_match(struct socket *clcsock,
+		       struct smc_clc_msg_proposal_prefix *prop);
+int smc_clc_wait_msg(struct smc_sock *smc, void *buf, int buflen,
+		     u8 expected_type, unsigned long timeout);
+int smc_clc_send_decline(struct smc_sock *smc, u32 peer_diag_info, u8 version);
+int smc_clc_send_proposal(struct smc_sock *smc, struct smc_init_info *ini);
+int smc_clc_send_confirm(struct smc_sock *smc, bool clnt_first_contact,
+			 u8 version, u8 *eid, struct smc_init_info *ini);
+int smc_clc_send_accept(struct smc_sock *smc, bool srv_first_contact,
+			u8 version, u8 *negotiated_eid, struct smc_init_info *ini);
+int smc_clc_srv_v2x_features_validate(struct smc_sock *smc,
+				      struct smc_clc_msg_proposal *pclc,
+				      struct smc_init_info *ini);
+int smc_clc_clnt_v2x_features_validate(struct smc_clc_first_contact_ext *fce,
+				       struct smc_init_info *ini);
+int smc_clc_v2x_features_confirm_check(struct smc_clc_msg_accept_confirm *cclc,
+				       struct smc_init_info *ini);
+void smc_clc_init(void) __init;
+void smc_clc_exit(void);
+void smc_clc_get_hostname(u8 **host);
+bool smc_clc_match_eid(u8 *negotiated_eid,
+		       struct smc_clc_v2_extension *smc_v2_ext,
+		       u8 *peer_eid, u8 *local_eid);
+int smc_clc_ueid_count(void);
+int smc_nl_dump_ueid(struct sk_buff *skb, struct netlink_callback *cb);
+int smc_nl_add_ueid(struct sk_buff *skb, struct genl_info *info);
+int smc_nl_remove_ueid(struct sk_buff *skb, struct genl_info *info);
+int smc_nl_flush_ueid(struct sk_buff *skb, struct genl_info *info);
+int smc_nl_dump_seid(struct sk_buff *skb, struct netlink_callback *cb);
+int smc_nl_enable_seid(struct sk_buff *skb, struct genl_info *info);
+int smc_nl_disable_seid(struct sk_buff *skb, struct genl_info *info);
+
+#endif

linux-7.2.3/net/smc/smc_close.c

@@ -0,0 +1,506 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Socket Closing - normal and abnormal
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#include <linux/workqueue.h>
+#include <linux/sched/signal.h>
+
+#include <net/sock.h>
+#include <net/tcp.h>
+
+#include "smc.h"
+#include "smc_tx.h"
+#include "smc_cdc.h"
+#include "smc_close.h"
+
+/* release the clcsock that is assigned to the smc_sock */
+void smc_clcsock_release(struct smc_sock *smc)
+{
+	struct socket *tcp;
+
+	if (smc->listen_smc && current_work() != &smc->smc_listen_work)
+		cancel_work_sync(&smc->smc_listen_work);
+	mutex_lock(&smc->clcsock_release_lock);
+	if (smc->clcsock) {
+		tcp = smc->clcsock;
+		smc->clcsock = NULL;
+		sock_release(tcp);
+	}
+	mutex_unlock(&smc->clcsock_release_lock);
+}
+
+static void smc_close_cleanup_listen(struct sock *parent)
+{
+	struct sock *sk;
+
+	/* Close non-accepted connections */
+	while ((sk = smc_accept_dequeue(parent, NULL)))
+		smc_close_non_accepted(sk);
+}
+
+/* wait for sndbuf data being transmitted */
+static void smc_close_stream_wait(struct smc_sock *smc, long timeout)
+{
+	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	struct sock *sk = &smc->sk;
+
+	if (!timeout)
+		return;
+
+	if (!smc_tx_prepared_sends(&smc->conn))
+		return;
+
+	/* Send out corked data remaining in sndbuf */
+	smc_tx_pending(&smc->conn);
+
+	smc->wait_close_tx_prepared = 1;
+	add_wait_queue(sk_sleep(sk), &wait);
+	while (!signal_pending(current) && timeout) {
+		int rc;
+
+		rc = sk_wait_event(sk, &timeout,
+				   !smc_tx_prepared_sends(&smc->conn) ||
+				   READ_ONCE(sk->sk_err) == ECONNABORTED ||
+				   READ_ONCE(sk->sk_err) == ECONNRESET ||
+				   smc->conn.killed,
+				   &wait);
+		if (rc)
+			break;
+	}
+	remove_wait_queue(sk_sleep(sk), &wait);
+	smc->wait_close_tx_prepared = 0;
+}
+
+void smc_close_wake_tx_prepared(struct smc_sock *smc)
+{
+	if (smc->wait_close_tx_prepared)
+		/* wake up socket closing */
+		smc->sk.sk_state_change(&smc->sk);
+}
+
+static int smc_close_wr(struct smc_connection *conn)
+{
+	conn->local_tx_ctrl.conn_state_flags.peer_done_writing = 1;
+
+	return smc_cdc_get_slot_and_msg_send(conn);
+}
+
+static int smc_close_final(struct smc_connection *conn)
+{
+	if (atomic_read(&conn->bytes_to_rcv))
+		conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
+	else
+		conn->local_tx_ctrl.conn_state_flags.peer_conn_closed = 1;
+	if (conn->killed)
+		return -EPIPE;
+
+	return smc_cdc_get_slot_and_msg_send(conn);
+}
+
+int smc_close_abort(struct smc_connection *conn)
+{
+	conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
+
+	return smc_cdc_get_slot_and_msg_send(conn);
+}
+
+static void smc_close_cancel_work(struct smc_sock *smc)
+{
+	struct sock *sk = &smc->sk;
+
+	release_sock(sk);
+	if (cancel_work_sync(&smc->conn.close_work))
+		sock_put(sk);
+	cancel_delayed_work_sync(&smc->conn.tx_work);
+	lock_sock(sk);
+}
+
+/* terminate smc socket abnormally - active abort
+ * link group is terminated, i.e. RDMA communication no longer possible
+ */
+void smc_close_active_abort(struct smc_sock *smc)
+{
+	struct sock *sk = &smc->sk;
+	bool release_clcsock = false;
+
+	if (sk->sk_state != SMC_INIT && smc->clcsock && smc->clcsock->sk) {
+		sk->sk_err = ECONNABORTED;
+		if (smc->clcsock && smc->clcsock->sk)
+			tcp_abort(smc->clcsock->sk, ECONNABORTED);
+	}
+	switch (sk->sk_state) {
+	case SMC_ACTIVE:
+	case SMC_APPCLOSEWAIT1:
+	case SMC_APPCLOSEWAIT2:
+		sk->sk_state = SMC_PEERABORTWAIT;
+		smc_close_cancel_work(smc);
+		if (sk->sk_state != SMC_PEERABORTWAIT)
+			break;
+		sk->sk_state = SMC_CLOSED;
+		sock_put(sk); /* (postponed) passive closing */
+		break;
+	case SMC_PEERCLOSEWAIT1:
+	case SMC_PEERCLOSEWAIT2:
+	case SMC_PEERFINCLOSEWAIT:
+		sk->sk_state = SMC_PEERABORTWAIT;
+		smc_close_cancel_work(smc);
+		if (sk->sk_state != SMC_PEERABORTWAIT)
+			break;
+		sk->sk_state = SMC_CLOSED;
+		smc_conn_free(&smc->conn);
+		release_clcsock = true;
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_PROCESSABORT:
+	case SMC_APPFINCLOSEWAIT:
+		sk->sk_state = SMC_PEERABORTWAIT;
+		smc_close_cancel_work(smc);
+		if (sk->sk_state != SMC_PEERABORTWAIT)
+			break;
+		sk->sk_state = SMC_CLOSED;
+		smc_conn_free(&smc->conn);
+		release_clcsock = true;
+		break;
+	case SMC_INIT:
+	case SMC_PEERABORTWAIT:
+	case SMC_CLOSED:
+		break;
+	}
+
+	smc_sock_set_flag(sk, SOCK_DEAD);
+	sk->sk_state_change(sk);
+
+	if (release_clcsock) {
+		release_sock(sk);
+		smc_clcsock_release(smc);
+		lock_sock(sk);
+	}
+}
+
+static inline bool smc_close_sent_any_close(struct smc_connection *conn)
+{
+	return conn->local_tx_ctrl.conn_state_flags.peer_conn_abort ||
+	       conn->local_tx_ctrl.conn_state_flags.peer_conn_closed;
+}
+
+int smc_close_active(struct smc_sock *smc)
+{
+	struct smc_cdc_conn_state_flags *txflags =
+		&smc->conn.local_tx_ctrl.conn_state_flags;
+	struct smc_connection *conn = &smc->conn;
+	struct sock *sk = &smc->sk;
+	int old_state;
+	long timeout;
+	int rc = 0;
+	int rc1 = 0;
+
+	timeout = current->flags & PF_EXITING ?
+		  0 : sock_flag(sk, SOCK_LINGER) ?
+		      sk->sk_lingertime : SMC_MAX_STREAM_WAIT_TIMEOUT;
+
+	old_state = sk->sk_state;
+again:
+	switch (sk->sk_state) {
+	case SMC_INIT:
+		sk->sk_state = SMC_CLOSED;
+		break;
+	case SMC_LISTEN:
+		sk->sk_state = SMC_CLOSED;
+		sk->sk_state_change(sk); /* wake up accept */
+		if (smc->clcsock && smc->clcsock->sk) {
+			write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
+			smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
+					       &smc->clcsk_data_ready);
+			rcu_assign_sk_user_data(smc->clcsock->sk, NULL);
+			write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
+			rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
+		}
+		smc_close_cleanup_listen(sk);
+		release_sock(sk);
+		flush_work(&smc->tcp_listen_work);
+		lock_sock(sk);
+		break;
+	case SMC_ACTIVE:
+		smc_close_stream_wait(smc, timeout);
+		release_sock(sk);
+		cancel_delayed_work_sync(&conn->tx_work);
+		lock_sock(sk);
+		if (sk->sk_state == SMC_ACTIVE) {
+			/* send close request */
+			rc = smc_close_final(conn);
+			sk->sk_state = SMC_PEERCLOSEWAIT1;
+
+			/* actively shutdown clcsock before peer close it,
+			 * prevent peer from entering TIME_WAIT state.
+			 */
+			if (smc->clcsock && smc->clcsock->sk) {
+				rc1 = kernel_sock_shutdown(smc->clcsock,
+							   SHUT_RDWR);
+				rc = rc ? rc : rc1;
+			}
+		} else {
+			/* peer event has changed the state */
+			goto again;
+		}
+		break;
+	case SMC_APPFINCLOSEWAIT:
+		/* socket already shutdown wr or both (active close) */
+		if (txflags->peer_done_writing &&
+		    !smc_close_sent_any_close(conn)) {
+			/* just shutdown wr done, send close request */
+			rc = smc_close_final(conn);
+		}
+		sk->sk_state = SMC_CLOSED;
+		break;
+	case SMC_APPCLOSEWAIT1:
+	case SMC_APPCLOSEWAIT2:
+		if (!smc_cdc_rxed_any_close(conn))
+			smc_close_stream_wait(smc, timeout);
+		release_sock(sk);
+		cancel_delayed_work_sync(&conn->tx_work);
+		lock_sock(sk);
+		if (sk->sk_state != SMC_APPCLOSEWAIT1 &&
+		    sk->sk_state != SMC_APPCLOSEWAIT2)
+			goto again;
+		/* confirm close from peer */
+		rc = smc_close_final(conn);
+		if (smc_cdc_rxed_any_close(conn)) {
+			/* peer has closed the socket already */
+			sk->sk_state = SMC_CLOSED;
+			sock_put(sk); /* postponed passive closing */
+		} else {
+			/* peer has just issued a shutdown write */
+			sk->sk_state = SMC_PEERFINCLOSEWAIT;
+		}
+		break;
+	case SMC_PEERCLOSEWAIT1:
+	case SMC_PEERCLOSEWAIT2:
+		if (txflags->peer_done_writing &&
+		    !smc_close_sent_any_close(conn)) {
+			/* just shutdown wr done, send close request */
+			rc = smc_close_final(conn);
+		}
+		/* peer sending PeerConnectionClosed will cause transition */
+		break;
+	case SMC_PEERFINCLOSEWAIT:
+		/* peer sending PeerConnectionClosed will cause transition */
+		break;
+	case SMC_PROCESSABORT:
+		rc = smc_close_abort(conn);
+		sk->sk_state = SMC_CLOSED;
+		break;
+	case SMC_PEERABORTWAIT:
+		sk->sk_state = SMC_CLOSED;
+		break;
+	case SMC_CLOSED:
+		/* nothing to do, add tracing in future patch */
+		break;
+	}
+
+	if (old_state != sk->sk_state)
+		sk->sk_state_change(sk);
+	return rc;
+}
+
+static void smc_close_passive_abort_received(struct smc_sock *smc)
+{
+	struct smc_cdc_conn_state_flags *txflags =
+		&smc->conn.local_tx_ctrl.conn_state_flags;
+	struct sock *sk = &smc->sk;
+
+	switch (sk->sk_state) {
+	case SMC_INIT:
+	case SMC_ACTIVE:
+	case SMC_APPCLOSEWAIT1:
+		sk->sk_state = SMC_PROCESSABORT;
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_APPFINCLOSEWAIT:
+		sk->sk_state = SMC_PROCESSABORT;
+		break;
+	case SMC_PEERCLOSEWAIT1:
+	case SMC_PEERCLOSEWAIT2:
+		if (txflags->peer_done_writing &&
+		    !smc_close_sent_any_close(&smc->conn))
+			/* just shutdown, but not yet closed locally */
+			sk->sk_state = SMC_PROCESSABORT;
+		else
+			sk->sk_state = SMC_CLOSED;
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_APPCLOSEWAIT2:
+	case SMC_PEERFINCLOSEWAIT:
+		sk->sk_state = SMC_CLOSED;
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_PEERABORTWAIT:
+		sk->sk_state = SMC_CLOSED;
+		break;
+	case SMC_PROCESSABORT:
+	/* nothing to do, add tracing in future patch */
+		break;
+	}
+}
+
+/* Either some kind of closing has been received: peer_conn_closed,
+ * peer_conn_abort, or peer_done_writing
+ * or the link group of the connection terminates abnormally.
+ */
+static void smc_close_passive_work(struct work_struct *work)
+{
+	struct smc_connection *conn = container_of(work,
+						   struct smc_connection,
+						   close_work);
+	struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
+	struct smc_cdc_conn_state_flags *rxflags;
+	bool release_clcsock = false;
+	struct sock *sk = &smc->sk;
+	int old_state;
+
+	lock_sock(sk);
+	old_state = sk->sk_state;
+
+	rxflags = &conn->local_rx_ctrl.conn_state_flags;
+	if (rxflags->peer_conn_abort) {
+		/* peer has not received all data */
+		smc_close_passive_abort_received(smc);
+		release_sock(sk);
+		cancel_delayed_work_sync(&conn->tx_work);
+		lock_sock(sk);
+		goto wakeup;
+	}
+
+	switch (sk->sk_state) {
+	case SMC_INIT:
+		sk->sk_state = SMC_APPCLOSEWAIT1;
+		break;
+	case SMC_ACTIVE:
+		sk->sk_state = SMC_APPCLOSEWAIT1;
+		/* postpone sock_put() for passive closing to cover
+		 * received SEND_SHUTDOWN as well
+		 */
+		break;
+	case SMC_PEERCLOSEWAIT1:
+		if (rxflags->peer_done_writing)
+			sk->sk_state = SMC_PEERCLOSEWAIT2;
+		fallthrough;
+		/* to check for closing */
+	case SMC_PEERCLOSEWAIT2:
+		if (!smc_cdc_rxed_any_close(conn))
+			break;
+		if (sock_flag(sk, SOCK_DEAD) &&
+		    smc_close_sent_any_close(conn)) {
+			/* smc_release has already been called locally */
+			sk->sk_state = SMC_CLOSED;
+		} else {
+			/* just shutdown, but not yet closed locally */
+			sk->sk_state = SMC_APPFINCLOSEWAIT;
+		}
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_PEERFINCLOSEWAIT:
+		if (smc_cdc_rxed_any_close(conn)) {
+			sk->sk_state = SMC_CLOSED;
+			sock_put(sk); /* passive closing */
+		}
+		break;
+	case SMC_APPCLOSEWAIT1:
+	case SMC_APPCLOSEWAIT2:
+		/* postpone sock_put() for passive closing to cover
+		 * received SEND_SHUTDOWN as well
+		 */
+		break;
+	case SMC_APPFINCLOSEWAIT:
+	case SMC_PEERABORTWAIT:
+	case SMC_PROCESSABORT:
+	case SMC_CLOSED:
+		/* nothing to do, add tracing in future patch */
+		break;
+	}
+
+wakeup:
+	sk->sk_data_ready(sk); /* wakeup blocked rcvbuf consumers */
+	sk->sk_write_space(sk); /* wakeup blocked sndbuf producers */
+
+	if (old_state != sk->sk_state) {
+		sk->sk_state_change(sk);
+		if ((sk->sk_state == SMC_CLOSED) &&
+		    (sock_flag(sk, SOCK_DEAD) || !sk->sk_socket)) {
+			smc_conn_free(conn);
+			if (smc->clcsock)
+				release_clcsock = true;
+		}
+	}
+	release_sock(sk);
+	if (release_clcsock)
+		smc_clcsock_release(smc);
+	sock_put(sk); /* sock_hold done by schedulers of close_work */
+}
+
+int smc_close_shutdown_write(struct smc_sock *smc)
+{
+	struct smc_connection *conn = &smc->conn;
+	struct sock *sk = &smc->sk;
+	int old_state;
+	long timeout;
+	int rc = 0;
+
+	timeout = current->flags & PF_EXITING ?
+		  0 : sock_flag(sk, SOCK_LINGER) ?
+		      sk->sk_lingertime : SMC_MAX_STREAM_WAIT_TIMEOUT;
+
+	old_state = sk->sk_state;
+again:
+	switch (sk->sk_state) {
+	case SMC_ACTIVE:
+		smc_close_stream_wait(smc, timeout);
+		release_sock(sk);
+		cancel_delayed_work_sync(&conn->tx_work);
+		lock_sock(sk);
+		if (sk->sk_state != SMC_ACTIVE)
+			goto again;
+		/* send close wr request */
+		rc = smc_close_wr(conn);
+		sk->sk_state = SMC_PEERCLOSEWAIT1;
+		break;
+	case SMC_APPCLOSEWAIT1:
+		/* passive close */
+		if (!smc_cdc_rxed_any_close(conn))
+			smc_close_stream_wait(smc, timeout);
+		release_sock(sk);
+		cancel_delayed_work_sync(&conn->tx_work);
+		lock_sock(sk);
+		if (sk->sk_state != SMC_APPCLOSEWAIT1)
+			goto again;
+		/* confirm close from peer */
+		rc = smc_close_wr(conn);
+		sk->sk_state = SMC_APPCLOSEWAIT2;
+		break;
+	case SMC_APPCLOSEWAIT2:
+	case SMC_PEERFINCLOSEWAIT:
+	case SMC_PEERCLOSEWAIT1:
+	case SMC_PEERCLOSEWAIT2:
+	case SMC_APPFINCLOSEWAIT:
+	case SMC_PROCESSABORT:
+	case SMC_PEERABORTWAIT:
+		/* nothing to do, add tracing in future patch */
+		break;
+	}
+
+	if (old_state != sk->sk_state)
+		sk->sk_state_change(sk);
+	return rc;
+}
+
+/* Initialize close properties on connection establishment. */
+void smc_close_init(struct smc_sock *smc)
+{
+	INIT_WORK(&smc->conn.close_work, smc_close_passive_work);
+}

linux-7.2.3/net/smc/smc_close.h

@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Socket Closing
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef SMC_CLOSE_H
+#define SMC_CLOSE_H
+
+#include <linux/workqueue.h>
+
+#include "smc.h"
+
+#define SMC_MAX_STREAM_WAIT_TIMEOUT		(2 * HZ)
+#define SMC_CLOSE_SOCK_PUT_DELAY		HZ
+
+void smc_close_wake_tx_prepared(struct smc_sock *smc);
+int smc_close_active(struct smc_sock *smc);
+int smc_close_shutdown_write(struct smc_sock *smc);
+void smc_close_init(struct smc_sock *smc);
+void smc_clcsock_release(struct smc_sock *smc);
+int smc_close_abort(struct smc_connection *conn);
+void smc_close_active_abort(struct smc_sock *smc);
+
+#endif /* SMC_CLOSE_H */

linux-7.2.3/net/smc/smc_core.h

@@ -0,0 +1,629 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for SMC Connections, Link Groups and Links
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef _SMC_CORE_H
+#define _SMC_CORE_H
+
+#include <linux/atomic.h>
+#include <linux/types.h>
+#include <linux/smc.h>
+#include <linux/pci.h>
+#include <rdma/ib_verbs.h>
+#include <net/genetlink.h>
+#include <net/smc.h>
+
+#include "smc.h"
+#include "smc_ib.h"
+#include "smc_clc.h"
+
+#define SMC_RMBS_PER_LGR_MAX	255	/* max. # of RMBs per link group */
+#define SMC_CONN_PER_LGR_MIN	16	/* min. # of connections per link group */
+#define SMC_CONN_PER_LGR_MAX	255	/* max. # of connections per link group,
+					 * also is the default value for SMC-R v1 and v2.0
+					 */
+#define SMC_CONN_PER_LGR_PREFER	255	/* Preferred connections per link group used for
+					 * SMC-R v2.1 and later negotiation, vendors or
+					 * distributions may modify it to a value between
+					 * 16-255 as needed.
+					 */
+#define SMCR_MAX_SEND_WR_DEF	16	/* Default number of work requests per send queue */
+#define SMCR_MAX_RECV_WR_DEF	48	/* Default number of work requests per recv queue */
+
+struct smc_lgr_list {			/* list of link group definition */
+	struct list_head	list;
+	spinlock_t		lock;	/* protects list of link groups */
+	u32			num;	/* unique link group number */
+};
+
+enum smc_lgr_role {		/* possible roles of a link group */
+	SMC_CLNT,	/* client */
+	SMC_SERV	/* server */
+};
+
+enum smc_link_state {			/* possible states of a link */
+	SMC_LNK_UNUSED,		/* link is unused */
+	SMC_LNK_INACTIVE,	/* link is inactive */
+	SMC_LNK_ACTIVATING,	/* link is being activated */
+	SMC_LNK_ACTIVE,		/* link is active */
+};
+
+#define SMC_WR_BUF_SIZE		48	/* size of work request buffer */
+#define SMC_WR_BUF_V2_SIZE	8192	/* size of v2 work request buffer */
+
+struct smc_wr_buf {
+	u8	raw[SMC_WR_BUF_SIZE];
+};
+
+struct smc_wr_v2_buf {
+	u8	raw[SMC_WR_BUF_V2_SIZE];
+};
+
+#define SMC_WR_REG_MR_WAIT_TIME	(5 * HZ)/* wait time for ib_wr_reg_mr result */
+
+enum smc_wr_reg_state {
+	POSTED,		/* ib_wr_reg_mr request posted */
+	CONFIRMED,	/* ib_wr_reg_mr response: successful */
+	FAILED		/* ib_wr_reg_mr response: failure */
+};
+
+struct smc_rdma_sge {				/* sges for RDMA writes */
+	struct ib_sge		wr_tx_rdma_sge[SMC_IB_MAX_SEND_SGE];
+};
+
+#define SMC_MAX_RDMA_WRITES	2		/* max. # of RDMA writes per
+						 * message send
+						 */
+
+struct smc_rdma_sges {				/* sges per message send */
+	struct smc_rdma_sge	tx_rdma_sge[SMC_MAX_RDMA_WRITES];
+};
+
+struct smc_rdma_wr {				/* work requests per message
+						 * send
+						 */
+	struct ib_rdma_wr	wr_tx_rdma[SMC_MAX_RDMA_WRITES];
+};
+
+#define SMC_LGR_ID_SIZE		4
+
+struct smc_link {
+	struct smc_ib_device	*smcibdev;	/* ib-device */
+	u8			ibport;		/* port - values 1 | 2 */
+	struct ib_pd		*roce_pd;	/* IB protection domain,
+						 * unique for every RoCE QP
+						 */
+	struct ib_qp		*roce_qp;	/* IB queue pair */
+	struct ib_qp_attr	qp_attr;	/* IB queue pair attributes */
+
+	struct smc_wr_buf	*wr_tx_bufs;	/* WR send payload buffers */
+	struct ib_send_wr	*wr_tx_ibs;	/* WR send meta data */
+	struct ib_sge		*wr_tx_sges;	/* WR send gather meta data */
+	struct smc_rdma_sges	*wr_tx_rdma_sges;/*RDMA WRITE gather meta data*/
+	struct smc_rdma_wr	*wr_tx_rdmas;	/* WR RDMA WRITE */
+	struct smc_wr_tx_pend	*wr_tx_pends;	/* WR send waiting for CQE */
+	struct completion	*wr_tx_compl;	/* WR send CQE completion */
+	/* above four vectors have wr_tx_cnt elements and use the same index */
+	struct ib_send_wr	*wr_tx_v2_ib;	/* WR send v2 meta data */
+	struct ib_sge		*wr_tx_v2_sge;	/* WR send v2 gather meta data*/
+	struct smc_wr_tx_pend	*wr_tx_v2_pend;	/* WR send v2 waiting for CQE */
+	dma_addr_t		wr_tx_dma_addr;	/* DMA address of wr_tx_bufs */
+	dma_addr_t		wr_tx_v2_dma_addr; /* DMA address of v2 tx buf*/
+	atomic_long_t		wr_tx_id;	/* seq # of last sent WR */
+	unsigned long		*wr_tx_mask;	/* bit mask of used indexes */
+	u32			wr_tx_cnt;	/* number of WR send buffers */
+	wait_queue_head_t	wr_tx_wait;	/* wait for free WR send buf */
+	struct {
+		struct percpu_ref	wr_tx_refs;
+	} ____cacheline_aligned_in_smp;
+	struct completion	tx_ref_comp;
+
+	u8			*wr_rx_bufs;	/* WR recv payload buffers */
+	struct ib_recv_wr	*wr_rx_ibs;	/* WR recv meta data */
+	struct ib_sge		*wr_rx_sges;	/* WR recv scatter meta data */
+	/* above three vectors have wr_rx_cnt elements and use the same index */
+	int			wr_rx_sge_cnt; /* rx sge, V1 is 1, V2 is either 2 or 1 */
+	int			wr_rx_buflen;	/* buffer len for the first sge, len for the
+						 * second sge is lgr shared if rx sge is 2.
+						 */
+	dma_addr_t		wr_rx_dma_addr;	/* DMA address of wr_rx_bufs */
+	dma_addr_t		wr_rx_v2_dma_addr; /* DMA address of v2 rx buf*/
+	u64			wr_rx_id;	/* seq # of last recv WR */
+	u64			wr_rx_id_compl; /* seq # of last completed WR */
+	u32			wr_rx_cnt;	/* number of WR recv buffers */
+	unsigned long		wr_rx_tstamp;	/* jiffies when last buf rx */
+	wait_queue_head_t       wr_rx_empty_wait; /* wait for RQ empty */
+
+	struct ib_reg_wr	wr_reg;		/* WR register memory region */
+	wait_queue_head_t	wr_reg_wait;	/* wait for wr_reg result */
+	struct {
+		struct percpu_ref	wr_reg_refs;
+	} ____cacheline_aligned_in_smp;
+	struct completion	reg_ref_comp;
+	enum smc_wr_reg_state	wr_reg_state;	/* state of wr_reg request */
+
+	u8			gid[SMC_GID_SIZE];/* gid matching used vlan id*/
+	u8			sgid_index;	/* gid index for vlan id      */
+	u32			peer_qpn;	/* QP number of peer */
+	enum ib_mtu		path_mtu;	/* used mtu */
+	enum ib_mtu		peer_mtu;	/* mtu size of peer */
+	u32			psn_initial;	/* QP tx initial packet seqno */
+	u32			peer_psn;	/* QP rx initial packet seqno */
+	u8			peer_mac[ETH_ALEN];	/* = gid[8:10||13:15] */
+	u8			peer_gid[SMC_GID_SIZE];	/* gid of peer*/
+	u8			link_id;	/* unique # within link group */
+	u8			link_uid[SMC_LGR_ID_SIZE]; /* unique lnk id */
+	u8			peer_link_uid[SMC_LGR_ID_SIZE]; /* peer uid */
+	u8			link_idx;	/* index in lgr link array */
+	u8			link_is_asym;	/* is link asymmetric? */
+	u8			clearing : 1;	/* link is being cleared */
+	refcount_t		refcnt;		/* link reference count */
+	struct smc_link_group	*lgr;		/* parent link group */
+	struct work_struct	link_down_wrk;	/* wrk to bring link down */
+	char			ibname[IB_DEVICE_NAME_MAX]; /* ib device name */
+	int			ndev_ifidx; /* network device ifindex */
+
+	enum smc_link_state	state;		/* state of link */
+	struct delayed_work	llc_testlink_wrk; /* testlink worker */
+	struct completion	llc_testlink_resp; /* wait for rx of testlink */
+	int			llc_testlink_time; /* testlink interval */
+	atomic_t		conn_cnt; /* connections on this link */
+	u16			max_send_wr;
+	u16			max_recv_wr;
+};
+
+/* For now we just allow one parallel link per link group. The SMC protocol
+ * allows more (up to 8).
+ */
+#define SMC_LINKS_PER_LGR_MAX	3
+#define SMC_SINGLE_LINK		0
+#define SMC_LINKS_ADD_LNK_MIN	1	/* min. # of links per link group */
+#define SMC_LINKS_ADD_LNK_MAX	2	/* max. # of links per link group, also is the
+					 * default value for smc-r v1.0 and v2.0
+					 */
+#define SMC_LINKS_PER_LGR_MAX_PREFER	2	/* Preferred max links per link group used for
+						 * SMC-R v2.1 and later negotiation, vendors or
+						 * distributions may modify it to a value between
+						 * 1-2 as needed.
+						 */
+
+/* tx/rx buffer list element for sndbufs list and rmbs list of a lgr */
+struct smc_buf_desc {
+	struct list_head	list;
+	void			*cpu_addr;	/* virtual address of buffer */
+	struct page		*pages;
+	int			len;		/* length of buffer */
+	u32			used;		/* currently used / unused */
+	union {
+		struct { /* SMC-R */
+			struct sg_table	sgt[SMC_LINKS_PER_LGR_MAX];
+					/* virtual buffer */
+			struct ib_mr	*mr[SMC_LINKS_PER_LGR_MAX];
+					/* memory region: for rmb and
+					 * vzalloced sndbuf
+					 * incl. rkey provided to peer
+					 * and lkey provided to local
+					 */
+			u32		order;	/* allocation order */
+
+			u8		is_conf_rkey;
+					/* confirm_rkey done */
+			u8		is_reg_mr[SMC_LINKS_PER_LGR_MAX];
+					/* mem region registered */
+			u8		is_map_ib[SMC_LINKS_PER_LGR_MAX];
+					/* mem region mapped to lnk */
+			u8		is_dma_need_sync;
+			u8		is_reg_err;
+					/* buffer registration err */
+			u8		is_vm;
+					/* virtually contiguous */
+		};
+		struct { /* SMC-D */
+			/* SMC-D tx buffer */
+			bool		is_attached;
+					/* no need for explicit writes */
+			 /* SMC-D rx buffer: */
+			unsigned short	sba_idx;
+					/* SBA index number */
+			u64		token;
+					/* DMB token number */
+			dma_addr_t	dma_addr;
+					/* DMA address */
+		};
+	};
+};
+
+struct smc_rtoken {				/* address/key of remote RMB */
+	u64			dma_addr;
+	u32			rkey;
+};
+
+#define SMC_BUF_MIN_SIZE	16384	/* minimum size of an RMB */
+#define SMC_RMBE_SIZES		16	/* number of distinct RMBE sizes */
+/* theoretically, the RFC states that largest size would be 512K,
+ * i.e. compressed 5 and thus 6 sizes (0..5), despite
+ * struct smc_clc_msg_accept_confirm.rmbe_size being a 4 bit value (0..15)
+ */
+
+struct smcd_dev;
+
+enum smc_lgr_type {				/* redundancy state of lgr */
+	SMC_LGR_NONE,			/* no active links, lgr to be deleted */
+	SMC_LGR_SINGLE,			/* 1 active RNIC on each peer */
+	SMC_LGR_SYMMETRIC,		/* 2 active RNICs on each peer */
+	SMC_LGR_ASYMMETRIC_PEER,	/* local has 2, peer 1 active RNICs */
+	SMC_LGR_ASYMMETRIC_LOCAL,	/* local has 1, peer 2 active RNICs */
+};
+
+enum smcr_buf_type {		/* types of SMC-R sndbufs and RMBs */
+	SMCR_PHYS_CONT_BUFS	= 0,
+	SMCR_VIRT_CONT_BUFS	= 1,
+	SMCR_MIXED_BUFS		= 2,
+};
+
+enum smc_llc_flowtype {
+	SMC_LLC_FLOW_NONE	= 0,
+	SMC_LLC_FLOW_ADD_LINK	= 2,
+	SMC_LLC_FLOW_DEL_LINK	= 4,
+	SMC_LLC_FLOW_REQ_ADD_LINK = 5,
+	SMC_LLC_FLOW_RKEY	= 6,
+};
+
+struct smc_llc_qentry;
+
+struct smc_llc_flow {
+	enum smc_llc_flowtype type;
+	struct smc_llc_qentry *qentry;
+};
+
+struct smc_link_group {
+	struct list_head	list;
+	struct rb_root		conns_all;	/* connection tree */
+	rwlock_t		conns_lock;	/* protects conns_all */
+	unsigned int		conns_num;	/* current # of connections */
+	unsigned short		vlan_id;	/* vlan id of link group */
+
+	struct list_head	sndbufs[SMC_RMBE_SIZES];/* tx buffers */
+	struct rw_semaphore	sndbufs_lock;	/* protects tx buffers */
+	struct list_head	rmbs[SMC_RMBE_SIZES];	/* rx buffers */
+	struct rw_semaphore	rmbs_lock;	/* protects rx buffers */
+	u64			alloc_sndbufs;	/* stats of tx buffers */
+	u64			alloc_rmbs;	/* stats of rx buffers */
+
+	u8			id[SMC_LGR_ID_SIZE];	/* unique lgr id */
+	struct delayed_work	free_work;	/* delayed freeing of an lgr */
+	struct work_struct	terminate_work;	/* abnormal lgr termination */
+	struct workqueue_struct	*tx_wq;		/* wq for conn. tx workers */
+	u8			sync_err : 1;	/* lgr no longer fits to peer */
+	u8			terminating : 1;/* lgr is terminating */
+	u8			freeing : 1;	/* lgr is being freed */
+
+	refcount_t		refcnt;		/* lgr reference count */
+	bool			is_smcd;	/* SMC-R or SMC-D */
+	u8			smc_version;
+	u8			negotiated_eid[SMC_MAX_EID_LEN];
+	u8			peer_os;	/* peer operating system */
+	u8			peer_smc_release;
+	u8			peer_hostname[SMC_MAX_HOSTNAME_LEN];
+	union {
+		struct { /* SMC-R */
+			enum smc_lgr_role	role;
+						/* client or server */
+			struct smc_link		lnk[SMC_LINKS_PER_LGR_MAX];
+						/* smc link */
+			struct smc_wr_v2_buf	*wr_rx_buf_v2;
+						/* WR v2 recv payload buffer */
+			struct smc_wr_v2_buf	*wr_tx_buf_v2;
+						/* WR v2 send payload buffer */
+			char			peer_systemid[SMC_SYSTEMID_LEN];
+						/* unique system_id of peer */
+			struct smc_rtoken	rtokens[SMC_RMBS_PER_LGR_MAX]
+						[SMC_LINKS_PER_LGR_MAX];
+						/* remote addr/key pairs */
+			DECLARE_BITMAP(rtokens_used_mask, SMC_RMBS_PER_LGR_MAX);
+						/* used rtoken elements */
+			u8			next_link_id;
+			enum smc_lgr_type	type;
+			enum smcr_buf_type	buf_type;
+						/* redundancy state */
+			u8			pnet_id[SMC_MAX_PNETID_LEN + 1];
+						/* pnet id of this lgr */
+			struct list_head	llc_event_q;
+						/* queue for llc events */
+			spinlock_t		llc_event_q_lock;
+						/* protects llc_event_q */
+			struct rw_semaphore	llc_conf_mutex;
+						/* protects lgr reconfig. */
+			struct work_struct	llc_add_link_work;
+			struct work_struct	llc_del_link_work;
+			struct work_struct	llc_event_work;
+						/* llc event worker */
+			wait_queue_head_t	llc_flow_waiter;
+						/* w4 next llc event */
+			wait_queue_head_t	llc_msg_waiter;
+						/* w4 next llc msg */
+			struct smc_llc_flow	llc_flow_lcl;
+						/* llc local control field */
+			struct smc_llc_flow	llc_flow_rmt;
+						/* llc remote control field */
+			struct smc_llc_qentry	*delayed_event;
+						/* arrived when flow active */
+			spinlock_t		llc_flow_lock;
+						/* protects llc flow */
+			int			llc_testlink_time;
+						/* link keep alive time */
+			u32			llc_termination_rsn;
+						/* rsn code for termination */
+			u8			nexthop_mac[ETH_ALEN];
+			u8			uses_gateway;
+			__be32			saddr;
+						/* net namespace */
+			struct net		*net;
+			u8			max_conns;
+						/* max conn can be assigned to lgr */
+			u8			max_links;
+						/* max links can be added in lgr */
+			u16			max_send_wr;
+						/* number of WR buffers on send */
+			u16			max_recv_wr;
+						/* number of WR buffers on recv */
+		};
+		struct { /* SMC-D */
+			struct smcd_gid		peer_gid;
+						/* Peer GID (remote) */
+			struct smcd_dev		*smcd;
+						/* ISM device for VLAN reg. */
+			u8			peer_shutdown : 1;
+						/* peer triggered shutdownn */
+		};
+	};
+};
+
+struct smc_clc_msg_local;
+
+#define GID_LIST_SIZE	2
+
+struct smc_gidlist {
+	u8			len;
+	u8			list[GID_LIST_SIZE][SMC_GID_SIZE];
+};
+
+struct smc_init_info_smcrv2 {
+	/* Input fields */
+	__be32			saddr;
+	struct sock		*clc_sk;
+	__be32			daddr;
+
+	/* Output fields when saddr is set */
+	struct smc_ib_device	*ib_dev_v2;
+	u8			ib_port_v2;
+	u8			ib_gid_v2[SMC_GID_SIZE];
+
+	/* Additional output fields when clc_sk and daddr is set as well */
+	u8			uses_gateway;
+	u8			nexthop_mac[ETH_ALEN];
+
+	struct smc_gidlist	gidlist;
+};
+
+#define SMC_MAX_V2_ISM_DEVS	SMCD_CLC_MAX_V2_GID_ENTRIES
+				/* max # of proposed non-native ISM devices,
+				 * which can't exceed the max # of CHID-GID
+				 * entries in CLC proposal SMC-Dv2 extension.
+				 */
+struct smc_init_info {
+	u8			is_smcd;
+	u8			smc_type_v1;
+	u8			smc_type_v2;
+	u8			release_nr;
+	u8			max_conns;
+	u8			max_links;
+	u8			first_contact_peer;
+	u8			first_contact_local;
+	u16			feature_mask;
+	unsigned short		vlan_id;
+	u32			rc;
+	u8			negotiated_eid[SMC_MAX_EID_LEN];
+	/* SMC-R */
+	u8			smcr_version;
+	u8			check_smcrv2;
+	u8			peer_gid[SMC_GID_SIZE];
+	u8			peer_mac[ETH_ALEN];
+	u8			peer_systemid[SMC_SYSTEMID_LEN];
+	struct smc_ib_device	*ib_dev;
+	u8			ib_gid[SMC_GID_SIZE];
+	u8			ib_port;
+	u32			ib_clcqpn;
+	struct smc_init_info_smcrv2 smcrv2;
+	/* SMC-D */
+	struct smcd_gid		ism_peer_gid[SMC_MAX_V2_ISM_DEVS + 1];
+	struct smcd_dev		*ism_dev[SMC_MAX_V2_ISM_DEVS + 1];
+	u16			ism_chid[SMC_MAX_V2_ISM_DEVS + 1];
+	u8			ism_offered_cnt; /* # of ISM devices offered */
+	u8			ism_selected;    /* index of selected ISM dev*/
+	u8			smcd_version;
+};
+
+/* Find the connection associated with the given alert token in the link group.
+ * To use rbtrees we have to implement our own search core.
+ * Requires @conns_lock
+ * @token	alert token to search for
+ * @lgr		 link group to search in
+ * Returns connection associated with token if found, NULL otherwise.
+ */
+static inline struct smc_connection *smc_lgr_find_conn(
+	u32 token, struct smc_link_group *lgr)
+{
+	struct smc_connection *res = NULL;
+	struct rb_node *node;
+
+	node = lgr->conns_all.rb_node;
+	while (node) {
+		struct smc_connection *cur = rb_entry(node,
+					struct smc_connection, alert_node);
+
+		if (cur->alert_token_local > token) {
+			node = node->rb_left;
+		} else {
+			if (cur->alert_token_local < token) {
+				node = node->rb_right;
+			} else {
+				res = cur;
+				break;
+			}
+		}
+	}
+
+	return res;
+}
+
+static inline bool smc_conn_lgr_valid(struct smc_connection *conn)
+{
+	return conn->lgr && conn->alert_token_local;
+}
+
+/*
+ * Returns true if the specified link is usable.
+ *
+ * usable means the link is ready to receive RDMA messages, map memory
+ * on the link, etc. This doesn't ensure we are able to send RDMA messages
+ * on this link, if sending RDMA messages is needed, use smc_link_sendable()
+ */
+static inline bool smc_link_usable(struct smc_link *lnk)
+{
+	if (lnk->state == SMC_LNK_UNUSED || lnk->state == SMC_LNK_INACTIVE)
+		return false;
+	return true;
+}
+
+/*
+ * Returns true if the specified link is ready to receive AND send RDMA
+ * messages.
+ *
+ * For the client side in first contact, the underlying QP may still in
+ * RESET or RTR when the link state is ACTIVATING, checks in smc_link_usable()
+ * is not strong enough. For those places that need to send any CDC or LLC
+ * messages, use smc_link_sendable(), otherwise, use smc_link_usable() instead
+ */
+static inline bool smc_link_sendable(struct smc_link *lnk)
+{
+	return smc_link_usable(lnk) &&
+		lnk->qp_attr.cur_qp_state == IB_QPS_RTS;
+}
+
+static inline bool smc_link_active(struct smc_link *lnk)
+{
+	return lnk->state == SMC_LNK_ACTIVE;
+}
+
+static inline bool smc_link_shared_v2_rxbuf(struct smc_link *lnk)
+{
+	return lnk->wr_rx_sge_cnt > 1;
+}
+
+static inline void smc_gid_be16_convert(__u8 *buf, u8 *gid_raw)
+{
+	sprintf(buf, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x",
+		be16_to_cpu(((__be16 *)gid_raw)[0]),
+		be16_to_cpu(((__be16 *)gid_raw)[1]),
+		be16_to_cpu(((__be16 *)gid_raw)[2]),
+		be16_to_cpu(((__be16 *)gid_raw)[3]),
+		be16_to_cpu(((__be16 *)gid_raw)[4]),
+		be16_to_cpu(((__be16 *)gid_raw)[5]),
+		be16_to_cpu(((__be16 *)gid_raw)[6]),
+		be16_to_cpu(((__be16 *)gid_raw)[7]));
+}
+
+struct smc_pci_dev {
+	__u32		pci_fid;
+	__u16		pci_pchid;
+	__u16		pci_vendor;
+	__u16		pci_device;
+	__u8		pci_id[SMC_PCI_ID_STR_LEN];
+};
+
+static inline void smc_set_pci_values(struct pci_dev *pci_dev,
+				      struct smc_pci_dev *smc_dev)
+{
+	smc_dev->pci_vendor = pci_dev->vendor;
+	smc_dev->pci_device = pci_dev->device;
+	snprintf(smc_dev->pci_id, sizeof(smc_dev->pci_id), "%s",
+		 pci_name(pci_dev));
+#if IS_ENABLED(CONFIG_S390)
+	{ /* Set s390 specific PCI information */
+	struct zpci_dev *zdev;
+
+	zdev = to_zpci(pci_dev);
+	smc_dev->pci_fid = zdev->fid;
+	smc_dev->pci_pchid = zdev->pchid;
+	}
+#endif
+}
+
+struct smc_sock;
+struct smc_clc_msg_accept_confirm;
+
+void smc_lgr_cleanup_early(struct smc_link_group *lgr);
+void smc_lgr_terminate_sched(struct smc_link_group *lgr);
+void smc_lgr_hold(struct smc_link_group *lgr);
+void smc_lgr_put(struct smc_link_group *lgr);
+void smcr_port_add(struct smc_ib_device *smcibdev, u8 ibport);
+void smcr_port_err(struct smc_ib_device *smcibdev, u8 ibport);
+void smc_smcd_terminate(struct smcd_dev *dev, struct smcd_gid *peer_gid,
+			unsigned short vlan);
+void smc_smcd_terminate_all(struct smcd_dev *dev);
+void smc_smcr_terminate_all(struct smc_ib_device *smcibdev);
+int smc_buf_create(struct smc_sock *smc, bool is_smcd);
+int smcd_buf_attach(struct smc_sock *smc);
+int smc_uncompress_bufsize(u8 compressed);
+int smc_rmb_rtoken_handling(struct smc_connection *conn, struct smc_link *link,
+			    struct smc_clc_msg_accept_confirm *clc);
+int smc_rtoken_add(struct smc_link *lnk, __be64 nw_vaddr, __be32 nw_rkey);
+int smc_rtoken_delete(struct smc_link *lnk, __be32 nw_rkey);
+void smc_rtoken_set(struct smc_link_group *lgr, int link_idx, int link_idx_new,
+		    __be32 nw_rkey_known, __be64 nw_vaddr, __be32 nw_rkey);
+void smc_rtoken_set2(struct smc_link_group *lgr, int rtok_idx, int link_id,
+		     __be64 nw_vaddr, __be32 nw_rkey);
+void smc_sndbuf_sync_sg_for_device(struct smc_connection *conn);
+void smc_rmb_sync_sg_for_cpu(struct smc_connection *conn);
+int smc_vlan_by_tcpsk(struct socket *clcsock, struct smc_init_info *ini);
+
+void smc_conn_free(struct smc_connection *conn);
+int smc_conn_create(struct smc_sock *smc, struct smc_init_info *ini);
+int smc_core_init(void);
+void smc_core_exit(void);
+
+int smcr_link_init(struct smc_link_group *lgr, struct smc_link *lnk,
+		   u8 link_idx, struct smc_init_info *ini);
+void smcr_link_clear(struct smc_link *lnk, bool log);
+void smcr_link_hold(struct smc_link *lnk);
+void smcr_link_put(struct smc_link *lnk);
+void smc_switch_link_and_count(struct smc_connection *conn,
+			       struct smc_link *to_lnk);
+int smcr_buf_map_lgr(struct smc_link *lnk);
+int smcr_buf_reg_lgr(struct smc_link *lnk);
+void smcr_lgr_set_type(struct smc_link_group *lgr, enum smc_lgr_type new_type);
+void smcr_lgr_set_type_asym(struct smc_link_group *lgr,
+			    enum smc_lgr_type new_type, int asym_lnk_idx);
+int smcr_link_reg_buf(struct smc_link *link, struct smc_buf_desc *rmb_desc);
+struct smc_link *smc_switch_conns(struct smc_link_group *lgr,
+				  struct smc_link *from_lnk, bool is_dev_err);
+void smcr_link_down_cond(struct smc_link *lnk);
+void smcr_link_down_cond_sched(struct smc_link *lnk);
+int smc_nl_get_sys_info(struct sk_buff *skb, struct netlink_callback *cb);
+int smcr_nl_get_lgr(struct sk_buff *skb, struct netlink_callback *cb);
+int smcr_nl_get_link(struct sk_buff *skb, struct netlink_callback *cb);
+int smcd_nl_get_lgr(struct sk_buff *skb, struct netlink_callback *cb);
+
+static inline struct smc_link_group *smc_get_lgr(struct smc_link *link)
+{
+	return link->lgr;
+}
+#endif

linux-7.2.3/net/smc/smc_diag.c

@@ -0,0 +1,278 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Monitoring SMC transport protocol sockets
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/sock_diag.h>
+#include <linux/inet_diag.h>
+#include <linux/smc_diag.h>
+#include <net/netlink.h>
+#include <net/smc.h>
+
+#include "smc.h"
+#include "smc_core.h"
+#include "smc_ism.h"
+
+struct smc_diag_dump_ctx {
+	int pos[2];
+};
+
+static struct smc_diag_dump_ctx *smc_dump_context(struct netlink_callback *cb)
+{
+	return (struct smc_diag_dump_ctx *)cb->ctx;
+}
+
+static void smc_diag_msg_common_fill(struct smc_diag_msg *r, struct sock *sk)
+{
+	struct smc_sock *smc = smc_sk(sk);
+
+	memset(r, 0, sizeof(*r));
+	r->diag_family = sk->sk_family;
+	sock_diag_save_cookie(sk, r->id.idiag_cookie);
+	if (!smc->clcsock)
+		return;
+	r->id.idiag_sport = htons(smc->clcsock->sk->sk_num);
+	r->id.idiag_dport = smc->clcsock->sk->sk_dport;
+	r->id.idiag_if = smc->clcsock->sk->sk_bound_dev_if;
+	if (sk->sk_protocol == SMCPROTO_SMC) {
+		r->id.idiag_src[0] = smc->clcsock->sk->sk_rcv_saddr;
+		r->id.idiag_dst[0] = smc->clcsock->sk->sk_daddr;
+#if IS_ENABLED(CONFIG_IPV6)
+	} else if (sk->sk_protocol == SMCPROTO_SMC6) {
+		memcpy(&r->id.idiag_src, &smc->clcsock->sk->sk_v6_rcv_saddr,
+		       sizeof(smc->clcsock->sk->sk_v6_rcv_saddr));
+		memcpy(&r->id.idiag_dst, &smc->clcsock->sk->sk_v6_daddr,
+		       sizeof(smc->clcsock->sk->sk_v6_daddr));
+#endif
+	}
+}
+
+static int smc_diag_msg_attrs_fill(struct sock *sk, struct sk_buff *skb,
+				   struct smc_diag_msg *r,
+				   struct user_namespace *user_ns)
+{
+	if (nla_put_u8(skb, SMC_DIAG_SHUTDOWN, sk->sk_shutdown))
+		return 1;
+
+	r->diag_uid = from_kuid_munged(user_ns, sk_uid(sk));
+	r->diag_inode = sock_i_ino(sk);
+	return 0;
+}
+
+static int __smc_diag_dump(struct sock *sk, struct sk_buff *skb,
+			   struct netlink_callback *cb,
+			   const struct smc_diag_req *req,
+			   struct nlattr *bc)
+{
+	struct smc_sock *smc = smc_sk(sk);
+	struct smc_diag_fallback fallback;
+	struct user_namespace *user_ns;
+	struct smc_diag_msg *r;
+	struct nlmsghdr *nlh;
+
+	nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			cb->nlh->nlmsg_type, sizeof(*r), NLM_F_MULTI);
+	if (!nlh)
+		return -EMSGSIZE;
+
+	r = nlmsg_data(nlh);
+	smc_diag_msg_common_fill(r, sk);
+	r->diag_state = sk->sk_state;
+	if (smc->use_fallback)
+		r->diag_mode = SMC_DIAG_MODE_FALLBACK_TCP;
+	else if (smc_conn_lgr_valid(&smc->conn) && smc->conn.lgr->is_smcd)
+		r->diag_mode = SMC_DIAG_MODE_SMCD;
+	else
+		r->diag_mode = SMC_DIAG_MODE_SMCR;
+	user_ns = sk_user_ns(NETLINK_CB(cb->skb).sk);
+	if (smc_diag_msg_attrs_fill(sk, skb, r, user_ns))
+		goto errout;
+
+	fallback.reason = smc->fallback_rsn;
+	fallback.peer_diagnosis = smc->peer_diagnosis;
+	if (nla_put(skb, SMC_DIAG_FALLBACK, sizeof(fallback), &fallback) < 0)
+		goto errout;
+
+	if ((req->diag_ext & (1 << (SMC_DIAG_CONNINFO - 1))) &&
+	    smc->conn.alert_token_local) {
+		struct smc_connection *conn = &smc->conn;
+		struct smc_diag_conninfo cinfo = {
+			.token = conn->alert_token_local,
+			.sndbuf_size = conn->sndbuf_desc ?
+				conn->sndbuf_desc->len : 0,
+			.rmbe_size = conn->rmb_desc ? conn->rmb_desc->len : 0,
+			.peer_rmbe_size = conn->peer_rmbe_size,
+
+			.rx_prod.wrap = conn->local_rx_ctrl.prod.wrap,
+			.rx_prod.count = conn->local_rx_ctrl.prod.count,
+			.rx_cons.wrap = conn->local_rx_ctrl.cons.wrap,
+			.rx_cons.count = conn->local_rx_ctrl.cons.count,
+
+			.tx_prod.wrap = conn->local_tx_ctrl.prod.wrap,
+			.tx_prod.count = conn->local_tx_ctrl.prod.count,
+			.tx_cons.wrap = conn->local_tx_ctrl.cons.wrap,
+			.tx_cons.count = conn->local_tx_ctrl.cons.count,
+
+			.tx_prod_flags =
+				*(u8 *)&conn->local_tx_ctrl.prod_flags,
+			.tx_conn_state_flags =
+				*(u8 *)&conn->local_tx_ctrl.conn_state_flags,
+			.rx_prod_flags = *(u8 *)&conn->local_rx_ctrl.prod_flags,
+			.rx_conn_state_flags =
+				*(u8 *)&conn->local_rx_ctrl.conn_state_flags,
+
+			.tx_prep.wrap = conn->tx_curs_prep.wrap,
+			.tx_prep.count = conn->tx_curs_prep.count,
+			.tx_sent.wrap = conn->tx_curs_sent.wrap,
+			.tx_sent.count = conn->tx_curs_sent.count,
+			.tx_fin.wrap = conn->tx_curs_fin.wrap,
+			.tx_fin.count = conn->tx_curs_fin.count,
+		};
+
+		if (nla_put(skb, SMC_DIAG_CONNINFO, sizeof(cinfo), &cinfo) < 0)
+			goto errout;
+	}
+
+	if (smc_conn_lgr_valid(&smc->conn) && !smc->conn.lgr->is_smcd &&
+	    (req->diag_ext & (1 << (SMC_DIAG_LGRINFO - 1))) &&
+	    !list_empty(&smc->conn.lgr->list)) {
+		struct smc_link *link = smc->conn.lnk;
+
+		struct smc_diag_lgrinfo linfo = {
+			.role = smc->conn.lgr->role,
+			.lnk[0].ibport = link->ibport,
+			.lnk[0].link_id = link->link_id,
+		};
+
+		memcpy(linfo.lnk[0].ibname, link->smcibdev->ibdev->name,
+		       sizeof(link->smcibdev->ibdev->name));
+		smc_gid_be16_convert(linfo.lnk[0].gid, link->gid);
+		smc_gid_be16_convert(linfo.lnk[0].peer_gid, link->peer_gid);
+
+		if (nla_put(skb, SMC_DIAG_LGRINFO, sizeof(linfo), &linfo) < 0)
+			goto errout;
+	}
+	if (smc_conn_lgr_valid(&smc->conn) && smc->conn.lgr->is_smcd &&
+	    (req->diag_ext & (1 << (SMC_DIAG_DMBINFO - 1))) &&
+	    !list_empty(&smc->conn.lgr->list) && smc->conn.rmb_desc) {
+		struct smc_connection *conn = &smc->conn;
+		struct smcd_diag_dmbinfo dinfo;
+		struct smcd_dev *smcd = conn->lgr->smcd;
+		struct smcd_gid smcd_gid;
+
+		memset(&dinfo, 0, sizeof(dinfo));
+
+		dinfo.linkid = *((u32 *)conn->lgr->id);
+		dinfo.peer_gid = conn->lgr->peer_gid.gid;
+		dinfo.peer_gid_ext = conn->lgr->peer_gid.gid_ext;
+		copy_to_smcdgid(&smcd_gid, &smcd->dibs->gid);
+		dinfo.my_gid = smcd_gid.gid;
+		dinfo.my_gid_ext = smcd_gid.gid_ext;
+		dinfo.token = conn->rmb_desc->token;
+		dinfo.peer_token = conn->peer_token;
+
+		if (nla_put(skb, SMC_DIAG_DMBINFO, sizeof(dinfo), &dinfo) < 0)
+			goto errout;
+	}
+
+	nlmsg_end(skb, nlh);
+	return 0;
+
+errout:
+	nlmsg_cancel(skb, nlh);
+	return -EMSGSIZE;
+}
+
+static int smc_diag_dump_proto(struct proto *prot, struct sk_buff *skb,
+			       struct netlink_callback *cb, int p_type)
+{
+	struct smc_diag_dump_ctx *cb_ctx = smc_dump_context(cb);
+	struct net *net = sock_net(skb->sk);
+	int snum = cb_ctx->pos[p_type];
+	struct nlattr *bc = NULL;
+	struct hlist_head *head;
+	int rc = 0, num = 0;
+	struct sock *sk;
+
+	read_lock(&prot->h.smc_hash->lock);
+	head = &prot->h.smc_hash->ht;
+	if (hlist_empty(head))
+		goto out;
+
+	sk_for_each(sk, head) {
+		if (!net_eq(sock_net(sk), net))
+			continue;
+		if (num < snum)
+			goto next;
+		rc = __smc_diag_dump(sk, skb, cb, nlmsg_data(cb->nlh), bc);
+		if (rc < 0)
+			goto out;
+next:
+		num++;
+	}
+
+out:
+	read_unlock(&prot->h.smc_hash->lock);
+	cb_ctx->pos[p_type] = num;
+	return rc;
+}
+
+static int smc_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	int rc = 0;
+
+	rc = smc_diag_dump_proto(&smc_proto, skb, cb, SMCPROTO_SMC);
+	if (!rc)
+		smc_diag_dump_proto(&smc_proto6, skb, cb, SMCPROTO_SMC6);
+	return skb->len;
+}
+
+static int smc_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h)
+{
+	struct net *net = sock_net(skb->sk);
+
+	if (h->nlmsg_type == SOCK_DIAG_BY_FAMILY &&
+	    h->nlmsg_flags & NLM_F_DUMP) {
+		{
+			struct netlink_dump_control c = {
+				.dump = smc_diag_dump,
+				.min_dump_alloc = SKB_WITH_OVERHEAD(32768),
+			};
+			return netlink_dump_start(net->diag_nlsk, skb, h, &c);
+		}
+	}
+	return 0;
+}
+
+static const struct sock_diag_handler smc_diag_handler = {
+	.owner = THIS_MODULE,
+	.family = AF_SMC,
+	.dump = smc_diag_handler_dump,
+};
+
+static int __init smc_diag_init(void)
+{
+	return sock_diag_register(&smc_diag_handler);
+}
+
+static void __exit smc_diag_exit(void)
+{
+	sock_diag_unregister(&smc_diag_handler);
+}
+
+module_init(smc_diag_init);
+module_exit(smc_diag_exit);
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("SMC socket monitoring via SOCK_DIAG");
+MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_NETLINK, NETLINK_SOCK_DIAG, 43 /* AF_SMC */);
+MODULE_ALIAS_GENL_FAMILY(SMCR_GENL_FAMILY_NAME);

linux-7.2.3/net/smc/smc_hs_bpf.c

@@ -0,0 +1,140 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Generic hook for SMC handshake flow.
+ *
+ *  Copyright IBM Corp. 2016
+ *  Copyright (c) 2025, Alibaba Inc.
+ *
+ *  Author: D. Wythe <alibuda@linux.alibaba.com>
+ */
+
+#include <linux/bpf_verifier.h>
+#include <linux/bpf.h>
+#include <linux/btf.h>
+#include <linux/rculist.h>
+
+#include "smc_hs_bpf.h"
+
+static DEFINE_SPINLOCK(smc_hs_ctrl_list_lock);
+static LIST_HEAD(smc_hs_ctrl_list);
+
+static int smc_hs_ctrl_reg(struct smc_hs_ctrl *ctrl)
+{
+	int ret = 0;
+
+	spin_lock(&smc_hs_ctrl_list_lock);
+	/* already exist or duplicate name */
+	if (smc_hs_ctrl_find_by_name(ctrl->name))
+		ret = -EEXIST;
+	else
+		list_add_tail_rcu(&ctrl->list, &smc_hs_ctrl_list);
+	spin_unlock(&smc_hs_ctrl_list_lock);
+	return ret;
+}
+
+static void smc_hs_ctrl_unreg(struct smc_hs_ctrl *ctrl)
+{
+	spin_lock(&smc_hs_ctrl_list_lock);
+	list_del_rcu(&ctrl->list);
+	spin_unlock(&smc_hs_ctrl_list_lock);
+
+	/* Ensure that all readers to complete */
+	synchronize_rcu();
+}
+
+struct smc_hs_ctrl *smc_hs_ctrl_find_by_name(const char *name)
+{
+	struct smc_hs_ctrl *ctrl;
+
+	list_for_each_entry_rcu(ctrl, &smc_hs_ctrl_list, list) {
+		if (strcmp(ctrl->name, name) == 0)
+			return ctrl;
+	}
+	return NULL;
+}
+
+static int __smc_bpf_stub_set_tcp_option(struct tcp_sock *tp) { return 1; }
+static int __smc_bpf_stub_set_tcp_option_cond(const struct tcp_sock *tp,
+					      struct inet_request_sock *ireq)
+{
+	return 1;
+}
+
+static struct smc_hs_ctrl __smc_bpf_hs_ctrl = {
+	.syn_option	= __smc_bpf_stub_set_tcp_option,
+	.synack_option	= __smc_bpf_stub_set_tcp_option_cond,
+};
+
+static int smc_bpf_hs_ctrl_init(struct btf *btf) { return 0; }
+
+static int smc_bpf_hs_ctrl_reg(void *kdata, struct bpf_link *link)
+{
+	if (link)
+		return -EOPNOTSUPP;
+
+	return smc_hs_ctrl_reg(kdata);
+}
+
+static void smc_bpf_hs_ctrl_unreg(void *kdata, struct bpf_link *link)
+{
+	smc_hs_ctrl_unreg(kdata);
+}
+
+static int smc_bpf_hs_ctrl_init_member(const struct btf_type *t,
+				       const struct btf_member *member,
+				       void *kdata, const void *udata)
+{
+	const struct smc_hs_ctrl *u_ctrl;
+	struct smc_hs_ctrl *k_ctrl;
+	u32 moff;
+
+	u_ctrl = (const struct smc_hs_ctrl *)udata;
+	k_ctrl = (struct smc_hs_ctrl *)kdata;
+
+	moff = __btf_member_bit_offset(t, member) / 8;
+	switch (moff) {
+	case offsetof(struct smc_hs_ctrl, name):
+		if (bpf_obj_name_cpy(k_ctrl->name, u_ctrl->name,
+				     sizeof(u_ctrl->name)) <= 0)
+			return -EINVAL;
+		return 1;
+	case offsetof(struct smc_hs_ctrl, flags):
+		if (u_ctrl->flags & ~SMC_HS_CTRL_ALL_FLAGS)
+			return -EINVAL;
+		k_ctrl->flags = u_ctrl->flags;
+		return 1;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static const struct bpf_func_proto *
+bpf_smc_hs_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
+{
+	return bpf_base_func_proto(func_id, prog);
+}
+
+static const struct bpf_verifier_ops smc_bpf_verifier_ops = {
+	.get_func_proto		= bpf_smc_hs_func_proto,
+	.is_valid_access	= bpf_tracing_btf_ctx_access,
+};
+
+static struct bpf_struct_ops bpf_smc_hs_ctrl_ops = {
+	.name		= "smc_hs_ctrl",
+	.init		= smc_bpf_hs_ctrl_init,
+	.reg		= smc_bpf_hs_ctrl_reg,
+	.unreg		= smc_bpf_hs_ctrl_unreg,
+	.cfi_stubs	= &__smc_bpf_hs_ctrl,
+	.verifier_ops	= &smc_bpf_verifier_ops,
+	.init_member	= smc_bpf_hs_ctrl_init_member,
+	.owner		= THIS_MODULE,
+};
+
+int bpf_smc_hs_ctrl_init(void)
+{
+	return register_bpf_struct_ops(&bpf_smc_hs_ctrl_ops, smc_hs_ctrl);
+}

linux-7.2.3/net/smc/smc_hs_bpf.h

@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Generic hook for SMC handshake flow.
+ *
+ *  Copyright IBM Corp. 2016
+ *  Copyright (c) 2025, Alibaba Inc.
+ *
+ *  Author: D. Wythe <alibuda@linux.alibaba.com>
+ */
+
+#ifndef __SMC_HS_CTRL
+#define __SMC_HS_CTRL
+
+#include <net/smc.h>
+
+/* Find hs_ctrl by the target name, which required to be a c-string.
+ * Return NULL if no such ctrl was found,otherwise, return a valid ctrl.
+ *
+ * Note: Caller MUST ensure it's was invoked under rcu_read_lock.
+ */
+struct smc_hs_ctrl *smc_hs_ctrl_find_by_name(const char *name);
+
+#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)
+int bpf_smc_hs_ctrl_init(void);
+#else
+static inline int bpf_smc_hs_ctrl_init(void) { return 0; }
+#endif /* CONFIG_SMC_HS_CTRL_BPF */
+
+#endif /* __SMC_HS_CTRL */

linux-7.2.3/net/smc/smc_ib.h

@@ -0,0 +1,119 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for IB environment
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Ursula Braun <Ursula Braun@linux.vnet.ibm.com>
+ */
+
+#ifndef _SMC_IB_H
+#define _SMC_IB_H
+
+#include <linux/interrupt.h>
+#include <linux/if_ether.h>
+#include <linux/mutex.h>
+#include <linux/wait.h>
+#include <rdma/ib_verbs.h>
+#include <net/smc.h>
+
+#define SMC_MAX_PORTS			2	/* Max # of ports */
+#define SMC_GID_SIZE			sizeof(union ib_gid)
+
+#define SMC_IB_MAX_SEND_SGE		2
+
+struct smc_ib_devices {			/* list of smc ib devices definition */
+	struct list_head	list;
+	struct mutex		mutex;	/* protects list of smc ib devices */
+};
+
+extern struct smc_ib_devices	smc_ib_devices; /* list of smc ib devices */
+extern struct smc_lgr_list smc_lgr_list; /* list of linkgroups */
+
+struct smc_ib_device {				/* ib-device infos for smc */
+	struct list_head	list;
+	struct ib_device	*ibdev;
+	struct ib_port_attr	pattr[SMC_MAX_PORTS];	/* ib dev. port attrs */
+	struct ib_event_handler	event_handler;	/* global ib_event handler */
+	struct ib_cq		*roce_cq_send;	/* send completion queue */
+	struct ib_cq		*roce_cq_recv;	/* recv completion queue */
+	struct tasklet_struct	send_tasklet;	/* called by send cq handler */
+	struct tasklet_struct	recv_tasklet;	/* called by recv cq handler */
+	char			mac[SMC_MAX_PORTS][ETH_ALEN];
+						/* mac address per port*/
+	u8			pnetid[SMC_MAX_PORTS][SMC_MAX_PNETID_LEN];
+						/* pnetid per port */
+	bool			pnetid_by_user[SMC_MAX_PORTS];
+						/* pnetid defined by user? */
+	u8			initialized : 1; /* ib dev CQ, evthdl done */
+	struct work_struct	port_event_work;
+	unsigned long		port_event_mask;
+	DECLARE_BITMAP(ports_going_away, SMC_MAX_PORTS);
+	atomic_t		lnk_cnt;	/* number of links on ibdev */
+	wait_queue_head_t	lnks_deleted;	/* wait 4 removal of all links*/
+	struct mutex		mutex;		/* protect dev setup+cleanup */
+	atomic_t		lnk_cnt_by_port[SMC_MAX_PORTS];
+						/* number of links per port */
+	int			ndev_ifidx[SMC_MAX_PORTS]; /* ndev if indexes */
+};
+
+static inline __be32 smc_ib_gid_to_ipv4(u8 gid[SMC_GID_SIZE])
+{
+	struct in6_addr *addr6 = (struct in6_addr *)gid;
+
+	if (ipv6_addr_v4mapped(addr6) ||
+	    !(addr6->s6_addr32[0] | addr6->s6_addr32[1] | addr6->s6_addr32[2]))
+		return addr6->s6_addr32[3];
+	return cpu_to_be32(INADDR_NONE);
+}
+
+static inline struct net *smc_ib_net(struct smc_ib_device *smcibdev)
+{
+	if (smcibdev && smcibdev->ibdev)
+		return read_pnet(&smcibdev->ibdev->coredev.rdma_net);
+	return NULL;
+}
+
+struct smc_init_info_smcrv2;
+struct smc_buf_desc;
+struct smc_link;
+
+void smc_ib_ndev_change(struct net_device *ndev, unsigned long event);
+int smc_ib_register_client(void) __init;
+void smc_ib_unregister_client(void);
+bool smc_ib_port_active(struct smc_ib_device *smcibdev, u8 ibport);
+int smc_ib_buf_map_sg(struct smc_link *lnk,
+		      struct smc_buf_desc *buf_slot,
+		      enum dma_data_direction data_direction);
+void smc_ib_buf_unmap_sg(struct smc_link *lnk,
+			 struct smc_buf_desc *buf_slot,
+			 enum dma_data_direction data_direction);
+void smc_ib_dealloc_protection_domain(struct smc_link *lnk);
+int smc_ib_create_protection_domain(struct smc_link *lnk);
+void smc_ib_destroy_queue_pair(struct smc_link *lnk);
+int smc_ib_create_queue_pair(struct smc_link *lnk);
+int smc_ib_ready_link(struct smc_link *lnk);
+int smc_ib_modify_qp_rts(struct smc_link *lnk);
+int smc_ib_modify_qp_error(struct smc_link *lnk);
+long smc_ib_setup_per_ibdev(struct smc_ib_device *smcibdev);
+int smc_ib_get_memory_region(struct ib_pd *pd, int access_flags,
+			     struct smc_buf_desc *buf_slot, u8 link_idx);
+void smc_ib_put_memory_region(struct ib_mr *mr);
+bool smc_ib_is_sg_need_sync(struct smc_link *lnk,
+			    struct smc_buf_desc *buf_slot);
+void smc_ib_sync_sg_for_cpu(struct smc_link *lnk,
+			    struct smc_buf_desc *buf_slot,
+			    enum dma_data_direction data_direction);
+void smc_ib_sync_sg_for_device(struct smc_link *lnk,
+			       struct smc_buf_desc *buf_slot,
+			       enum dma_data_direction data_direction);
+int smc_ib_determine_gid(struct smc_ib_device *smcibdev, u8 ibport,
+			 unsigned short vlan_id, u8 gid[], u8 *sgid_index,
+			 struct smc_init_info_smcrv2 *smcrv2);
+int smc_ib_find_route(struct net *net, __be32 saddr, __be32 daddr,
+		      u8 nexthop_mac[], u8 *uses_gateway);
+bool smc_ib_is_valid_local_systemid(void);
+int smcr_nl_get_device(struct sk_buff *skb, struct netlink_callback *cb);
+#endif

linux-7.2.3/net/smc/smc_inet.c

@@ -0,0 +1,163 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for the IPPROTO_SMC (socket related)
+ *
+ *  Copyright IBM Corp. 2016, 2018
+ *  Copyright (c) 2024, Alibaba Inc.
+ *
+ *  Author: D. Wythe <alibuda@linux.alibaba.com>
+ */
+
+#include <net/protocol.h>
+#include <net/sock.h>
+
+#include "smc_inet.h"
+#include "smc.h"
+
+static int smc_inet_init_sock(struct sock *sk);
+
+static struct proto smc_inet_prot = {
+	.name		= "INET_SMC",
+	.owner		= THIS_MODULE,
+	.init		= smc_inet_init_sock,
+	.hash		= smc_hash_sk,
+	.unhash		= smc_unhash_sk,
+	.release_cb	= smc_release_cb,
+	.obj_size	= sizeof(struct smc_sock),
+	.h.smc_hash	= &smc_v4_hashinfo,
+	.slab_flags	= SLAB_TYPESAFE_BY_RCU,
+};
+
+static const struct proto_ops smc_inet_stream_ops = {
+	.family		= PF_INET,
+	.owner		= THIS_MODULE,
+	.release	= smc_release,
+	.bind		= smc_bind,
+	.connect	= smc_connect,
+	.socketpair	= sock_no_socketpair,
+	.accept		= smc_accept,
+	.getname	= smc_getname,
+	.poll		= smc_poll,
+	.ioctl		= smc_ioctl,
+	.listen		= smc_listen,
+	.shutdown	= smc_shutdown,
+	.setsockopt	= smc_setsockopt,
+	.getsockopt	= smc_getsockopt,
+	.sendmsg	= smc_sendmsg,
+	.recvmsg	= smc_recvmsg,
+	.mmap		= sock_no_mmap,
+	.splice_read	= smc_splice_read,
+};
+
+static struct inet_protosw smc_inet_protosw = {
+	.type		= SOCK_STREAM,
+	.protocol	= IPPROTO_SMC,
+	.prot		= &smc_inet_prot,
+	.ops		= &smc_inet_stream_ops,
+};
+
+#if IS_ENABLED(CONFIG_IPV6)
+struct smc6_sock {
+	struct smc_sock		smc;
+	struct ipv6_pinfo	inet6;
+};
+
+static struct proto smc_inet6_prot = {
+	.name		= "INET6_SMC",
+	.owner		= THIS_MODULE,
+	.init		= smc_inet_init_sock,
+	.hash		= smc_hash_sk,
+	.unhash		= smc_unhash_sk,
+	.release_cb	= smc_release_cb,
+	.obj_size	= sizeof(struct smc6_sock),
+	.h.smc_hash	= &smc_v6_hashinfo,
+	.slab_flags	= SLAB_TYPESAFE_BY_RCU,
+	.ipv6_pinfo_offset	= offsetof(struct smc6_sock, inet6),
+};
+
+static const struct proto_ops smc_inet6_stream_ops = {
+	.family		= PF_INET6,
+	.owner		= THIS_MODULE,
+	.release	= smc_release,
+	.bind		= smc_bind,
+	.connect	= smc_connect,
+	.socketpair	= sock_no_socketpair,
+	.accept		= smc_accept,
+	.getname	= smc_getname,
+	.poll		= smc_poll,
+	.ioctl		= smc_ioctl,
+	.listen		= smc_listen,
+	.shutdown	= smc_shutdown,
+	.setsockopt	= smc_setsockopt,
+	.getsockopt	= smc_getsockopt,
+	.sendmsg	= smc_sendmsg,
+	.recvmsg	= smc_recvmsg,
+	.mmap		= sock_no_mmap,
+	.splice_read	= smc_splice_read,
+};
+
+static struct inet_protosw smc_inet6_protosw = {
+	.type		= SOCK_STREAM,
+	.protocol	= IPPROTO_SMC,
+	.prot		= &smc_inet6_prot,
+	.ops		= &smc_inet6_stream_ops,
+};
+#endif /* CONFIG_IPV6 */
+
+static int smc_inet_init_sock(struct sock *sk)
+{
+	struct net *net = sock_net(sk);
+
+	/* init common smc sock */
+	smc_sk_init(net, sk, IPPROTO_SMC);
+	/* create clcsock */
+	return smc_create_clcsk(net, sk, sk->sk_family);
+}
+
+int __init smc_inet_init(void)
+{
+	int rc;
+
+	rc = proto_register(&smc_inet_prot, 1);
+	if (rc) {
+		pr_err("%s: proto_register smc_inet_prot fails with %d\n",
+		       __func__, rc);
+		return rc;
+	}
+	/* no return value */
+	inet_register_protosw(&smc_inet_protosw);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	rc = proto_register(&smc_inet6_prot, 1);
+	if (rc) {
+		pr_err("%s: proto_register smc_inet6_prot fails with %d\n",
+		       __func__, rc);
+		goto out_inet6_prot;
+	}
+	rc = inet6_register_protosw(&smc_inet6_protosw);
+	if (rc) {
+		pr_err("%s: inet6_register_protosw smc_inet6_protosw fails with %d\n",
+		       __func__, rc);
+		goto out_inet6_protosw;
+	}
+	return rc;
+out_inet6_protosw:
+	proto_unregister(&smc_inet6_prot);
+out_inet6_prot:
+	inet_unregister_protosw(&smc_inet_protosw);
+	proto_unregister(&smc_inet_prot);
+#endif /* CONFIG_IPV6 */
+	return rc;
+}
+
+void smc_inet_exit(void)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+	inet6_unregister_protosw(&smc_inet6_protosw);
+	proto_unregister(&smc_inet6_prot);
+#endif /* CONFIG_IPV6 */
+	inet_unregister_protosw(&smc_inet_protosw);
+	proto_unregister(&smc_inet_prot);
+}

linux-7.2.3/net/smc/smc_inet.h

@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for the IPPROTO_SMC (socket related)
+
+ *  Copyright IBM Corp. 2016
+ *  Copyright (c) 2024, Alibaba Inc.
+ *
+ *  Author: D. Wythe <alibuda@linux.alibaba.com>
+ */
+#ifndef __INET_SMC
+#define __INET_SMC
+
+/* Initialize protocol registration on IPPROTO_SMC,
+ * @return 0 on success
+ */
+int smc_inet_init(void);
+
+void smc_inet_exit(void);
+
+#endif /* __INET_SMC */

linux-7.2.3/net/smc/smc_ism.c

@@ -0,0 +1,644 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Shared Memory Communications Direct over ISM devices (SMC-D)
+ *
+ * Functions for ISM device.
+ *
+ * Copyright IBM Corp. 2018
+ */
+
+#include <linux/if_vlan.h>
+#include <linux/spinlock.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <asm/page.h>
+
+#include "smc.h"
+#include "smc_core.h"
+#include "smc_ism.h"
+#include "smc_pnet.h"
+#include "smc_netlink.h"
+#include "linux/dibs.h"
+
+struct smcd_dev_list smcd_dev_list = {
+	.list = LIST_HEAD_INIT(smcd_dev_list.list),
+	.mutex = __MUTEX_INITIALIZER(smcd_dev_list.mutex)
+};
+
+static bool smc_ism_v2_capable;
+static u8 smc_ism_v2_system_eid[SMC_MAX_EID_LEN];
+
+static void smcd_register_dev(struct dibs_dev *dibs);
+static void smcd_unregister_dev(struct dibs_dev *dibs);
+static void smcd_handle_event(struct dibs_dev *dibs,
+			      const struct dibs_event *event);
+static void smcd_handle_irq(struct dibs_dev *dibs, unsigned int dmbno,
+			    u16 dmbemask);
+
+static struct dibs_client_ops smc_client_ops = {
+	.add_dev = smcd_register_dev,
+	.del_dev = smcd_unregister_dev,
+	.handle_event = smcd_handle_event,
+	.handle_irq = smcd_handle_irq,
+};
+
+static struct dibs_client smc_dibs_client = {
+	.name = "SMC-D",
+	.ops = &smc_client_ops,
+};
+
+static void smc_ism_create_system_eid(void)
+{
+	struct smc_ism_seid *seid =
+		(struct smc_ism_seid *)smc_ism_v2_system_eid;
+#if IS_ENABLED(CONFIG_S390)
+	struct cpuid id;
+	u16 ident_tail;
+	char tmp[5];
+
+	memcpy(seid->seid_string, "IBM-SYSZ-ISMSEID00000000", 24);
+	get_cpu_id(&id);
+	ident_tail = (u16)(id.ident & SMC_ISM_IDENT_MASK);
+	snprintf(tmp, 5, "%04X", ident_tail);
+	memcpy(seid->serial_number, tmp, 4);
+	snprintf(tmp, 5, "%04X", id.machine);
+	memcpy(seid->type, tmp, 4);
+#else
+	memset(seid, 0, SMC_MAX_EID_LEN);
+#endif
+}
+
+/* Test if an ISM communication is possible - same CPC */
+int smc_ism_cantalk(struct smcd_gid *peer_gid, unsigned short vlan_id,
+		    struct smcd_dev *smcd)
+{
+	struct dibs_dev *dibs = smcd->dibs;
+	uuid_t ism_rgid;
+
+	copy_to_dibsgid(&ism_rgid, peer_gid);
+	return dibs->ops->query_remote_gid(dibs, &ism_rgid, vlan_id ? 1 : 0,
+					  vlan_id);
+}
+
+void smc_ism_get_system_eid(u8 **eid)
+{
+	if (!smc_ism_v2_capable)
+		*eid = NULL;
+	else
+		*eid = smc_ism_v2_system_eid;
+}
+
+u16 smc_ism_get_chid(struct smcd_dev *smcd)
+{
+	return smcd->dibs->ops->get_fabric_id(smcd->dibs);
+}
+
+/* HW supports ISM V2 and thus System EID is defined */
+bool smc_ism_is_v2_capable(void)
+{
+	return smc_ism_v2_capable;
+}
+
+void smc_ism_set_v2_capable(void)
+{
+	smc_ism_v2_capable = true;
+}
+
+/* Set a connection using this DMBE. */
+void smc_ism_set_conn(struct smc_connection *conn)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&conn->lgr->smcd->lock, flags);
+	conn->lgr->smcd->conn[conn->rmb_desc->sba_idx] = conn;
+	spin_unlock_irqrestore(&conn->lgr->smcd->lock, flags);
+}
+
+/* Unset a connection using this DMBE. */
+void smc_ism_unset_conn(struct smc_connection *conn)
+{
+	unsigned long flags;
+
+	if (!conn->rmb_desc)
+		return;
+
+	spin_lock_irqsave(&conn->lgr->smcd->lock, flags);
+	conn->lgr->smcd->conn[conn->rmb_desc->sba_idx] = NULL;
+	spin_unlock_irqrestore(&conn->lgr->smcd->lock, flags);
+}
+
+/* Register a VLAN identifier with the ISM device. Use a reference count
+ * and add a VLAN identifier only when the first DMB using this VLAN is
+ * registered.
+ */
+int smc_ism_get_vlan(struct smcd_dev *smcd, unsigned short vlanid)
+{
+	struct smc_ism_vlanid *new_vlan, *vlan;
+	unsigned long flags;
+	int rc = 0;
+
+	if (!vlanid)			/* No valid vlan id */
+		return -EINVAL;
+	if (!smcd->dibs->ops->add_vlan_id)
+		return -EOPNOTSUPP;
+
+	/* create new vlan entry, in case we need it */
+	new_vlan = kzalloc_obj(*new_vlan);
+	if (!new_vlan)
+		return -ENOMEM;
+	new_vlan->vlanid = vlanid;
+	refcount_set(&new_vlan->refcnt, 1);
+
+	/* if there is an existing entry, increase count and return */
+	spin_lock_irqsave(&smcd->lock, flags);
+	list_for_each_entry(vlan, &smcd->vlan, list) {
+		if (vlan->vlanid == vlanid) {
+			refcount_inc(&vlan->refcnt);
+			kfree(new_vlan);
+			goto out;
+		}
+	}
+
+	/* no existing entry found.
+	 * add new entry to device; might fail, e.g., if HW limit reached
+	 */
+	if (smcd->dibs->ops->add_vlan_id(smcd->dibs, vlanid)) {
+		kfree(new_vlan);
+		rc = -EIO;
+		goto out;
+	}
+	list_add_tail(&new_vlan->list, &smcd->vlan);
+out:
+	spin_unlock_irqrestore(&smcd->lock, flags);
+	return rc;
+}
+
+/* Unregister a VLAN identifier with the ISM device. Use a reference count
+ * and remove a VLAN identifier only when the last DMB using this VLAN is
+ * unregistered.
+ */
+int smc_ism_put_vlan(struct smcd_dev *smcd, unsigned short vlanid)
+{
+	struct smc_ism_vlanid *vlan;
+	unsigned long flags;
+	bool found = false;
+	int rc = 0;
+
+	if (!vlanid)			/* No valid vlan id */
+		return -EINVAL;
+	if (!smcd->dibs->ops->del_vlan_id)
+		return -EOPNOTSUPP;
+
+	spin_lock_irqsave(&smcd->lock, flags);
+	list_for_each_entry(vlan, &smcd->vlan, list) {
+		if (vlan->vlanid == vlanid) {
+			if (!refcount_dec_and_test(&vlan->refcnt))
+				goto out;
+			found = true;
+			break;
+		}
+	}
+	if (!found) {
+		rc = -ENOENT;
+		goto out;		/* VLAN id not in table */
+	}
+
+	/* Found and the last reference just gone */
+	if (smcd->dibs->ops->del_vlan_id(smcd->dibs, vlanid))
+		rc = -EIO;
+	list_del(&vlan->list);
+	kfree(vlan);
+out:
+	spin_unlock_irqrestore(&smcd->lock, flags);
+	return rc;
+}
+
+void smc_ism_unregister_dmb(struct smcd_dev *smcd,
+			    struct smc_buf_desc *dmb_desc)
+{
+	struct dibs_dmb dmb;
+
+	if (!dmb_desc->dma_addr)
+		return;
+
+	memset(&dmb, 0, sizeof(dmb));
+	dmb.dmb_tok = dmb_desc->token;
+	dmb.idx = dmb_desc->sba_idx;
+	dmb.cpu_addr = dmb_desc->cpu_addr;
+	dmb.dma_addr = dmb_desc->dma_addr;
+	dmb.dmb_len = dmb_desc->len;
+
+	smcd->dibs->ops->unregister_dmb(smcd->dibs, &dmb);
+
+	return;
+}
+
+int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
+			 struct smc_buf_desc *dmb_desc)
+{
+	struct dibs_dev *dibs;
+	struct dibs_dmb dmb;
+	int rc;
+
+	memset(&dmb, 0, sizeof(dmb));
+	dmb.dmb_len = dmb_len;
+	dmb.idx = dmb_desc->sba_idx;
+	dmb.vlan_id = lgr->vlan_id;
+	copy_to_dibsgid(&dmb.rgid, &lgr->peer_gid);
+
+	dibs = lgr->smcd->dibs;
+	rc = dibs->ops->register_dmb(dibs, &dmb, &smc_dibs_client);
+	if (!rc) {
+		dmb_desc->sba_idx = dmb.idx;
+		dmb_desc->token = dmb.dmb_tok;
+		dmb_desc->cpu_addr = dmb.cpu_addr;
+		dmb_desc->dma_addr = dmb.dma_addr;
+		dmb_desc->len = dmb.dmb_len;
+	}
+	return rc;
+}
+
+bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd)
+{
+	/* for now only loopback-ism supports
+	 * merging sndbuf with peer DMB to avoid
+	 * data copies between them.
+	 */
+	return (smcd->dibs->ops->support_mmapped_rdmb &&
+		smcd->dibs->ops->support_mmapped_rdmb(smcd->dibs));
+}
+
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
+		       struct smc_buf_desc *dmb_desc)
+{
+	struct dibs_dmb dmb;
+	int rc = 0;
+
+	if (!dev->dibs->ops->attach_dmb)
+		return -EINVAL;
+
+	memset(&dmb, 0, sizeof(dmb));
+	dmb.dmb_tok = token;
+	rc = dev->dibs->ops->attach_dmb(dev->dibs, &dmb);
+	if (!rc) {
+		dmb_desc->sba_idx = dmb.idx;
+		dmb_desc->token = dmb.dmb_tok;
+		dmb_desc->cpu_addr = dmb.cpu_addr;
+		dmb_desc->dma_addr = dmb.dma_addr;
+		dmb_desc->len = dmb.dmb_len;
+		dmb_desc->is_attached = true;
+	}
+	return rc;
+}
+
+int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token)
+{
+	if (!dev->dibs->ops->detach_dmb)
+		return -EINVAL;
+
+	return dev->dibs->ops->detach_dmb(dev->dibs, token);
+}
+
+static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
+				  struct sk_buff *skb,
+				  struct netlink_callback *cb)
+{
+	char smc_pnet[SMC_MAX_PNETID_LEN + 1];
+	struct smc_pci_dev smc_pci_dev;
+	struct nlattr *port_attrs;
+	struct dibs_dev *dibs;
+	struct nlattr *attrs;
+	int use_cnt = 0;
+	void *nlh;
+
+	dibs = smcd->dibs;
+	nlh = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			  &smc_gen_nl_family, NLM_F_MULTI,
+			  SMC_NETLINK_GET_DEV_SMCD);
+	if (!nlh)
+		goto errmsg;
+	attrs = nla_nest_start(skb, SMC_GEN_DEV_SMCD);
+	if (!attrs)
+		goto errout;
+	use_cnt = atomic_read(&smcd->lgr_cnt);
+	if (nla_put_u32(skb, SMC_NLA_DEV_USE_CNT, use_cnt))
+		goto errattr;
+	if (nla_put_u8(skb, SMC_NLA_DEV_IS_CRIT, use_cnt > 0))
+		goto errattr;
+	memset(&smc_pci_dev, 0, sizeof(smc_pci_dev));
+	smc_set_pci_values(to_pci_dev(dibs->dev.parent), &smc_pci_dev);
+	if (nla_put_u32(skb, SMC_NLA_DEV_PCI_FID, smc_pci_dev.pci_fid))
+		goto errattr;
+	if (nla_put_u16(skb, SMC_NLA_DEV_PCI_CHID, smc_pci_dev.pci_pchid))
+		goto errattr;
+	if (nla_put_u16(skb, SMC_NLA_DEV_PCI_VENDOR, smc_pci_dev.pci_vendor))
+		goto errattr;
+	if (nla_put_u16(skb, SMC_NLA_DEV_PCI_DEVICE, smc_pci_dev.pci_device))
+		goto errattr;
+	if (nla_put_string(skb, SMC_NLA_DEV_PCI_ID, smc_pci_dev.pci_id))
+		goto errattr;
+
+	port_attrs = nla_nest_start(skb, SMC_NLA_DEV_PORT);
+	if (!port_attrs)
+		goto errattr;
+	if (nla_put_u8(skb, SMC_NLA_DEV_PORT_PNET_USR, smcd->pnetid_by_user))
+		goto errportattr;
+	memcpy(smc_pnet, smcd->pnetid, SMC_MAX_PNETID_LEN);
+	smc_pnet[SMC_MAX_PNETID_LEN] = 0;
+	if (nla_put_string(skb, SMC_NLA_DEV_PORT_PNETID, smc_pnet))
+		goto errportattr;
+
+	nla_nest_end(skb, port_attrs);
+	nla_nest_end(skb, attrs);
+	genlmsg_end(skb, nlh);
+	return 0;
+
+errportattr:
+	nla_nest_cancel(skb, port_attrs);
+errattr:
+	nla_nest_cancel(skb, attrs);
+errout:
+	nlmsg_cancel(skb, nlh);
+errmsg:
+	return -EMSGSIZE;
+}
+
+static void smc_nl_prep_smcd_dev(struct smcd_dev_list *dev_list,
+				 struct sk_buff *skb,
+				 struct netlink_callback *cb)
+{
+	struct smc_nl_dmp_ctx *cb_ctx = smc_nl_dmp_ctx(cb);
+	int snum = cb_ctx->pos[0];
+	struct smcd_dev *smcd;
+	int num = 0;
+
+	mutex_lock(&dev_list->mutex);
+	list_for_each_entry(smcd, &dev_list->list, list) {
+		if (num < snum)
+			goto next;
+		if (smc_ism_is_loopback(smcd->dibs))
+			goto next;
+		if (smc_nl_handle_smcd_dev(smcd, skb, cb))
+			goto errout;
+next:
+		num++;
+	}
+errout:
+	mutex_unlock(&dev_list->mutex);
+	cb_ctx->pos[0] = num;
+}
+
+int smcd_nl_get_device(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	smc_nl_prep_smcd_dev(&smcd_dev_list, skb, cb);
+	return skb->len;
+}
+
+struct smc_ism_event_work {
+	struct work_struct work;
+	struct smcd_dev *smcd;
+	struct dibs_event event;
+};
+
+#define ISM_EVENT_REQUEST		0x0001
+#define ISM_EVENT_RESPONSE		0x0002
+#define ISM_EVENT_REQUEST_IR		0x00000001
+#define ISM_EVENT_CODE_SHUTDOWN		0x80
+#define ISM_EVENT_CODE_TESTLINK		0x83
+
+union smcd_sw_event_info {
+	u64	info;
+	struct {
+		u8		uid[SMC_LGR_ID_SIZE];
+		unsigned short	vlan_id;
+		u16		code;
+	};
+};
+
+static void smcd_handle_sw_event(struct smc_ism_event_work *wrk)
+{
+	struct dibs_dev *dibs = wrk->smcd->dibs;
+	union smcd_sw_event_info ev_info;
+	struct smcd_gid peer_gid;
+	uuid_t ism_rgid;
+
+	copy_to_smcdgid(&peer_gid, &wrk->event.gid);
+	ev_info.info = wrk->event.data;
+	switch (wrk->event.subtype) {
+	case ISM_EVENT_CODE_SHUTDOWN:	/* Peer shut down DMBs */
+		smc_smcd_terminate(wrk->smcd, &peer_gid, ev_info.vlan_id);
+		break;
+	case ISM_EVENT_CODE_TESTLINK:	/* Activity timer */
+		if (ev_info.code == ISM_EVENT_REQUEST &&
+		    dibs->ops->signal_event) {
+			ev_info.code = ISM_EVENT_RESPONSE;
+			copy_to_dibsgid(&ism_rgid, &peer_gid);
+			dibs->ops->signal_event(dibs, &ism_rgid,
+					       ISM_EVENT_REQUEST_IR,
+					       ISM_EVENT_CODE_TESTLINK,
+					       ev_info.info);
+		}
+		break;
+	}
+}
+
+/* worker for SMC-D events */
+static void smc_ism_event_work(struct work_struct *work)
+{
+	struct smc_ism_event_work *wrk =
+		container_of(work, struct smc_ism_event_work, work);
+	struct smcd_gid smcd_gid;
+
+	copy_to_smcdgid(&smcd_gid, &wrk->event.gid);
+
+	switch (wrk->event.type) {
+	case DIBS_DEV_EVENT: /* GID event, token is peer GID */
+		smc_smcd_terminate(wrk->smcd, &smcd_gid, VLAN_VID_MASK);
+		break;
+	case DIBS_BUF_EVENT:
+		break;
+	case DIBS_SW_EVENT: /* Software defined event */
+		smcd_handle_sw_event(wrk);
+		break;
+	}
+	kfree(wrk);
+}
+
+static struct smcd_dev *smcd_alloc_dev(const char *name, int max_dmbs)
+{
+	struct smcd_dev *smcd;
+
+	smcd = kzalloc_flex(*smcd, conn, max_dmbs);
+	if (!smcd)
+		return NULL;
+
+	smcd->event_wq = alloc_ordered_workqueue("ism_evt_wq-%s)",
+						 WQ_MEM_RECLAIM, name);
+	if (!smcd->event_wq)
+		goto free_smcd;
+
+	spin_lock_init(&smcd->lock);
+	spin_lock_init(&smcd->lgr_lock);
+	INIT_LIST_HEAD(&smcd->vlan);
+	INIT_LIST_HEAD(&smcd->lgr_list);
+	init_waitqueue_head(&smcd->lgrs_deleted);
+	return smcd;
+
+free_smcd:
+	kfree(smcd);
+	return NULL;
+}
+
+static void smcd_register_dev(struct dibs_dev *dibs)
+{
+	struct smcd_dev *smcd, *fentry;
+	int max_dmbs;
+
+	max_dmbs = dibs->ops->max_dmbs();
+
+	smcd = smcd_alloc_dev(dev_name(&dibs->dev), max_dmbs);
+	if (!smcd)
+		return;
+
+	smcd->dibs = dibs;
+	dibs_set_priv(dibs, &smc_dibs_client, smcd);
+
+	if (smc_pnetid_by_dev_port(dibs->dev.parent, 0, smcd->pnetid))
+		smc_pnetid_by_table_smcd(smcd);
+
+	if (smc_ism_is_loopback(dibs) ||
+	    (dibs->ops->add_vlan_id &&
+	     !dibs->ops->add_vlan_id(dibs, ISM_RESERVED_VLANID))) {
+		smc_ism_set_v2_capable();
+	}
+
+	mutex_lock(&smcd_dev_list.mutex);
+	/* sort list:
+	 * - devices without pnetid before devices with pnetid;
+	 * - loopback-ism always at the very beginning;
+	 */
+	if (!smcd->pnetid[0]) {
+		fentry = list_first_entry_or_null(&smcd_dev_list.list,
+						  struct smcd_dev, list);
+		if (fentry && smc_ism_is_loopback(fentry->dibs))
+			list_add(&smcd->list, &fentry->list);
+		else
+			list_add(&smcd->list, &smcd_dev_list.list);
+	} else {
+		list_add_tail(&smcd->list, &smcd_dev_list.list);
+	}
+	mutex_unlock(&smcd_dev_list.mutex);
+
+	if (smc_pnet_is_pnetid_set(smcd->pnetid))
+		pr_warn_ratelimited("smc: adding smcd device %s with pnetid %.16s%s\n",
+				    dev_name(&dibs->dev), smcd->pnetid,
+				    smcd->pnetid_by_user ?
+					" (user defined)" :
+					"");
+	else
+		pr_warn_ratelimited("smc: adding smcd device %s without pnetid\n",
+				    dev_name(&dibs->dev));
+	return;
+}
+
+static void smcd_unregister_dev(struct dibs_dev *dibs)
+{
+	struct smcd_dev *smcd = dibs_get_priv(dibs, &smc_dibs_client);
+
+	pr_warn_ratelimited("smc: removing smcd device %s\n",
+			    dev_name(&dibs->dev));
+	smcd->going_away = 1;
+	smc_smcd_terminate_all(smcd);
+	mutex_lock(&smcd_dev_list.mutex);
+	list_del_init(&smcd->list);
+	mutex_unlock(&smcd_dev_list.mutex);
+	destroy_workqueue(smcd->event_wq);
+	kfree(smcd);
+}
+
+/* SMCD Device event handler. Called from ISM device interrupt handler.
+ * Parameters are ism device pointer,
+ * - event->type (0 --> DMB, 1 --> GID),
+ * - event->code (event code),
+ * - event->tok (either DMB token when event type 0, or GID when event type 1)
+ * - event->time (time of day)
+ * - event->info (debug info).
+ *
+ * Context:
+ * - Function called in IRQ context from ISM device driver event handler.
+ */
+static void smcd_handle_event(struct dibs_dev *dibs,
+			      const struct dibs_event *event)
+{
+	struct smcd_dev *smcd = dibs_get_priv(dibs, &smc_dibs_client);
+	struct smc_ism_event_work *wrk;
+
+	if (smcd->going_away)
+		return;
+	/* copy event to event work queue, and let it be handled there */
+	wrk = kmalloc_obj(*wrk, GFP_ATOMIC);
+	if (!wrk)
+		return;
+	INIT_WORK(&wrk->work, smc_ism_event_work);
+	wrk->smcd = smcd;
+	wrk->event = *event;
+	queue_work(smcd->event_wq, &wrk->work);
+}
+
+/* SMCD Device interrupt handler. Called from ISM device interrupt handler.
+ * Parameters are the ism device pointer, DMB number, and the DMBE bitmask.
+ * Find the connection and schedule the tasklet for this connection.
+ *
+ * Context:
+ * - Function called in IRQ context from ISM device driver IRQ handler.
+ */
+static void smcd_handle_irq(struct dibs_dev *dibs, unsigned int dmbno,
+			    u16 dmbemask)
+{
+	struct smcd_dev *smcd = dibs_get_priv(dibs, &smc_dibs_client);
+	struct smc_connection *conn = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&smcd->lock, flags);
+	conn = smcd->conn[dmbno];
+	if (conn && !conn->killed)
+		tasklet_schedule(&conn->rx_tsklet);
+	spin_unlock_irqrestore(&smcd->lock, flags);
+}
+
+int smc_ism_signal_shutdown(struct smc_link_group *lgr)
+{
+	int rc = 0;
+	union smcd_sw_event_info ev_info;
+	uuid_t ism_rgid;
+
+	if (lgr->peer_shutdown)
+		return 0;
+	if (!lgr->smcd->dibs->ops->signal_event)
+		return 0;
+
+	memcpy(ev_info.uid, lgr->id, SMC_LGR_ID_SIZE);
+	ev_info.vlan_id = lgr->vlan_id;
+	ev_info.code = ISM_EVENT_REQUEST;
+	copy_to_dibsgid(&ism_rgid, &lgr->peer_gid);
+	rc = lgr->smcd->dibs->ops->signal_event(lgr->smcd->dibs, &ism_rgid,
+					  ISM_EVENT_REQUEST_IR,
+					  ISM_EVENT_CODE_SHUTDOWN,
+					  ev_info.info);
+	return rc;
+}
+
+int smc_ism_init(void)
+{
+	int rc = 0;
+
+	smc_ism_v2_capable = false;
+	smc_ism_create_system_eid();
+
+	rc = dibs_register_client(&smc_dibs_client);
+	return rc;
+}
+
+void smc_ism_exit(void)
+{
+	dibs_unregister_client(&smc_dibs_client);
+}

linux-7.2.3/net/smc/smc_ism.h

@@ -0,0 +1,123 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Shared Memory Communications Direct over ISM devices (SMC-D)
+ *
+ * SMC-D ISM device structure definitions.
+ *
+ * Copyright IBM Corp. 2018
+ */
+
+#ifndef SMCD_ISM_H
+#define SMCD_ISM_H
+
+#include <linux/uio.h>
+#include <linux/types.h>
+#include <linux/mutex.h>
+#include <linux/dibs.h>
+
+#include "smc.h"
+
+#define SMC_EMULATED_ISM_CHID_MASK	0xFF00
+#define SMC_ISM_IDENT_MASK		0x00FFFF
+
+struct smcd_dev_list {	/* List of SMCD devices */
+	struct list_head list;
+	struct mutex mutex;	/* Protects list of devices */
+};
+
+extern struct smcd_dev_list	smcd_dev_list;	/* list of smcd devices */
+
+struct smc_ism_vlanid {			/* VLAN id set on ISM device */
+	struct list_head list;
+	unsigned short vlanid;		/* Vlan id */
+	refcount_t refcnt;		/* Reference count */
+};
+
+struct smc_ism_seid {
+	u8 seid_string[24];
+	u8 serial_number[4];
+	u8 type[4];
+};
+
+struct smcd_dev;
+
+int smc_ism_cantalk(struct smcd_gid *peer_gid, unsigned short vlan_id,
+		    struct smcd_dev *dev);
+void smc_ism_set_conn(struct smc_connection *conn);
+void smc_ism_unset_conn(struct smc_connection *conn);
+int smc_ism_get_vlan(struct smcd_dev *dev, unsigned short vlan_id);
+int smc_ism_put_vlan(struct smcd_dev *dev, unsigned short vlan_id);
+int smc_ism_register_dmb(struct smc_link_group *lgr, int buf_size,
+			 struct smc_buf_desc *dmb_desc);
+void smc_ism_unregister_dmb(struct smcd_dev *dev,
+			    struct smc_buf_desc *dmb_desc);
+bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd);
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
+		       struct smc_buf_desc *dmb_desc);
+int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token);
+int smc_ism_signal_shutdown(struct smc_link_group *lgr);
+void smc_ism_get_system_eid(u8 **eid);
+u16 smc_ism_get_chid(struct smcd_dev *dev);
+bool smc_ism_is_v2_capable(void);
+void smc_ism_set_v2_capable(void);
+int smc_ism_init(void);
+void smc_ism_exit(void);
+int smcd_nl_get_device(struct sk_buff *skb, struct netlink_callback *cb);
+
+static inline int smc_ism_write(struct smcd_dev *smcd, u64 dmb_tok,
+				unsigned int idx, bool sf, unsigned int offset,
+				void *data, size_t len)
+{
+	int rc;
+
+	rc = smcd->dibs->ops->move_data(smcd->dibs, dmb_tok, idx, sf, offset,
+				       data, len);
+
+	return rc < 0 ? rc : 0;
+}
+
+static inline bool __smc_ism_is_emulated(u16 chid)
+{
+	/* CHIDs in range of 0xFF00 to 0xFFFF are reserved
+	 * for Emulated-ISM device.
+	 *
+	 * loopback-ism:	0xFFFF
+	 * virtio-ism:		0xFF00 ~ 0xFFFE
+	 */
+	return ((chid & 0xFF00) == 0xFF00);
+}
+
+static inline bool smc_ism_is_emulated(struct smcd_dev *smcd)
+{
+	u16 chid = smcd->dibs->ops->get_fabric_id(smcd->dibs);
+
+	return __smc_ism_is_emulated(chid);
+}
+
+static inline bool smc_ism_is_loopback(struct dibs_dev *dibs)
+{
+	return (dibs->ops->get_fabric_id(dibs) == DIBS_LOOPBACK_FABRIC);
+}
+
+static inline void copy_to_smcdgid(struct smcd_gid *sgid, uuid_t *dibs_gid)
+{
+	__be64 temp;
+
+	memcpy(&temp, dibs_gid, sizeof(sgid->gid));
+	sgid->gid = ntohll(temp);
+	memcpy(&temp, (uint8_t *)dibs_gid + sizeof(sgid->gid),
+	       sizeof(sgid->gid_ext));
+	sgid->gid_ext = ntohll(temp);
+}
+
+static inline void copy_to_dibsgid(uuid_t *dibs_gid, struct smcd_gid *sgid)
+{
+	__be64 temp;
+
+	temp = htonll(sgid->gid);
+	memcpy(dibs_gid, &temp, sizeof(sgid->gid));
+	temp = htonll(sgid->gid_ext);
+	memcpy((uint8_t *)dibs_gid + sizeof(sgid->gid), &temp,
+	       sizeof(sgid->gid_ext));
+}
+
+#endif

linux-7.2.3/net/smc/smc_llc.h

@@ -0,0 +1,120 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for LLC (link layer control) message handling
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Klaus Wacker <Klaus.Wacker@de.ibm.com>
+ *              Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef SMC_LLC_H
+#define SMC_LLC_H
+
+#include "smc_wr.h"
+
+#define SMC_LLC_FLAG_RESP		0x80
+
+#define SMC_LLC_WAIT_FIRST_TIME		(5 * HZ)
+#define SMC_LLC_WAIT_TIME		(2 * HZ)
+#define SMC_LLC_TESTLINK_DEFAULT_TIME	(30 * HZ)
+
+enum smc_llc_reqresp {
+	SMC_LLC_REQ,
+	SMC_LLC_RESP
+};
+
+enum smc_llc_msg_type {
+	SMC_LLC_CONFIRM_LINK		= 0x01,
+	SMC_LLC_ADD_LINK		= 0x02,
+	SMC_LLC_ADD_LINK_CONT		= 0x03,
+	SMC_LLC_DELETE_LINK		= 0x04,
+	SMC_LLC_REQ_ADD_LINK		= 0x05,
+	SMC_LLC_CONFIRM_RKEY		= 0x06,
+	SMC_LLC_TEST_LINK		= 0x07,
+	SMC_LLC_CONFIRM_RKEY_CONT	= 0x08,
+	SMC_LLC_DELETE_RKEY		= 0x09,
+	/* V2 types */
+	SMC_LLC_CONFIRM_LINK_V2		= 0x21,
+	SMC_LLC_ADD_LINK_V2		= 0x22,
+	SMC_LLC_DELETE_LINK_V2		= 0x24,
+	SMC_LLC_REQ_ADD_LINK_V2		= 0x25,
+	SMC_LLC_CONFIRM_RKEY_V2		= 0x26,
+	SMC_LLC_TEST_LINK_V2		= 0x27,
+	SMC_LLC_DELETE_RKEY_V2		= 0x29,
+};
+
+#define smc_link_downing(state) \
+	(cmpxchg(state, SMC_LNK_ACTIVE, SMC_LNK_INACTIVE) == SMC_LNK_ACTIVE)
+
+/* LLC DELETE LINK Request Reason Codes */
+#define SMC_LLC_DEL_LOST_PATH		0x00010000
+#define SMC_LLC_DEL_OP_INIT_TERM	0x00020000
+#define SMC_LLC_DEL_PROG_INIT_TERM	0x00030000
+#define SMC_LLC_DEL_PROT_VIOL		0x00040000
+#define SMC_LLC_DEL_NO_ASYM_NEEDED	0x00050000
+/* LLC DELETE LINK Response Reason Codes */
+#define SMC_LLC_DEL_NOLNK	0x00100000  /* Unknown Link ID (no link) */
+#define SMC_LLC_DEL_NOLGR	0x00200000  /* Unknown Link Group */
+
+/* returns a usable link of the link group, or NULL */
+static inline struct smc_link *smc_llc_usable_link(struct smc_link_group *lgr)
+{
+	int i;
+
+	for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++)
+		if (smc_link_usable(&lgr->lnk[i]))
+			return &lgr->lnk[i];
+	return NULL;
+}
+
+/* set the termination reason code for the link group */
+static inline void smc_llc_set_termination_rsn(struct smc_link_group *lgr,
+					       u32 rsn)
+{
+	if (!lgr->llc_termination_rsn)
+		lgr->llc_termination_rsn = rsn;
+}
+
+/* transmit */
+int smc_llc_send_confirm_link(struct smc_link *lnk,
+			      enum smc_llc_reqresp reqresp);
+int smc_llc_send_add_link(struct smc_link *link, u8 mac[], u8 gid[],
+			  struct smc_link *link_new,
+			  enum smc_llc_reqresp reqresp);
+int smc_llc_send_delete_link(struct smc_link *link, u8 link_del_id,
+			     enum smc_llc_reqresp reqresp, bool orderly,
+			     u32 reason);
+void smc_llc_srv_delete_link_local(struct smc_link *link, u8 del_link_id);
+void smc_llc_lgr_init(struct smc_link_group *lgr, struct smc_sock *smc);
+void smc_llc_lgr_clear(struct smc_link_group *lgr);
+int smc_llc_link_init(struct smc_link *link);
+void smc_llc_link_active(struct smc_link *link);
+void smc_llc_link_clear(struct smc_link *link, bool log);
+int smc_llc_do_confirm_rkey(struct smc_link *send_link,
+			    struct smc_buf_desc *rmb_desc);
+int smc_llc_do_delete_rkey(struct smc_link_group *lgr,
+			   struct smc_buf_desc *rmb_desc);
+int smc_llc_flow_initiate(struct smc_link_group *lgr,
+			  enum smc_llc_flowtype type);
+void smc_llc_flow_stop(struct smc_link_group *lgr, struct smc_llc_flow *flow);
+int smc_llc_eval_conf_link(struct smc_llc_qentry *qentry,
+			   enum smc_llc_reqresp type);
+void smc_llc_link_set_uid(struct smc_link *link);
+void smc_llc_save_peer_uid(struct smc_llc_qentry *qentry);
+struct smc_llc_qentry *smc_llc_wait(struct smc_link_group *lgr,
+				    struct smc_link *lnk,
+				    int time_out, u8 exp_msg);
+struct smc_llc_qentry *smc_llc_flow_qentry_clr(struct smc_llc_flow *flow);
+void smc_llc_flow_qentry_del(struct smc_llc_flow *flow);
+void smc_llc_send_link_delete_all(struct smc_link_group *lgr, bool ord,
+				  u32 rsn);
+int smc_llc_cli_add_link(struct smc_link *link, struct smc_llc_qentry *qentry);
+int smc_llc_srv_add_link(struct smc_link *link,
+			 struct smc_llc_qentry *req_qentry);
+void smc_llc_add_link_local(struct smc_link *link);
+int smc_llc_init(void) __init;
+
+#endif /* SMC_LLC_H */

linux-7.2.3/net/smc/smc_netlink.c

@@ -0,0 +1,157 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Generic netlink support functions to interact with SMC module
+ *
+ *  Copyright IBM Corp. 2020
+ *
+ *  Author(s):	Guvenc Gulce <guvenc@linux.ibm.com>
+ */
+
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/ctype.h>
+#include <linux/mutex.h>
+#include <linux/if.h>
+#include <linux/smc.h>
+
+#include "smc_core.h"
+#include "smc_ism.h"
+#include "smc_ib.h"
+#include "smc_clc.h"
+#include "smc_stats.h"
+#include "smc_netlink.h"
+
+const struct nla_policy
+smc_gen_ueid_policy[SMC_NLA_EID_TABLE_MAX + 1] = {
+	[SMC_NLA_EID_TABLE_UNSPEC]	= { .type = NLA_UNSPEC },
+	[SMC_NLA_EID_TABLE_ENTRY]	= { .type = NLA_STRING,
+					    .len = SMC_MAX_EID_LEN,
+					  },
+};
+
+#define SMC_CMD_MAX_ATTR 1
+/* SMC_GENL generic netlink operation definition */
+static const struct genl_ops smc_gen_nl_ops[] = {
+	{
+		.cmd = SMC_NETLINK_GET_SYS_INFO,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smc_nl_get_sys_info,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_LGR_SMCR,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smcr_nl_get_lgr,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_LINK_SMCR,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smcr_nl_get_link,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_LGR_SMCD,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smcd_nl_get_lgr,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_DEV_SMCD,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smcd_nl_get_device,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_DEV_SMCR,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smcr_nl_get_device,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_STATS,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smc_nl_get_stats,
+	},
+	{
+		.cmd = SMC_NETLINK_GET_FBACK_STATS,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smc_nl_get_fback_stats,
+	},
+	{
+		.cmd = SMC_NETLINK_DUMP_UEID,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smc_nl_dump_ueid,
+	},
+	{
+		.cmd = SMC_NETLINK_ADD_UEID,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_add_ueid,
+		.policy = smc_gen_ueid_policy,
+	},
+	{
+		.cmd = SMC_NETLINK_REMOVE_UEID,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_remove_ueid,
+		.policy = smc_gen_ueid_policy,
+	},
+	{
+		.cmd = SMC_NETLINK_FLUSH_UEID,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_flush_ueid,
+	},
+	{
+		.cmd = SMC_NETLINK_DUMP_SEID,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smc_nl_dump_seid,
+	},
+	{
+		.cmd = SMC_NETLINK_ENABLE_SEID,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_enable_seid,
+	},
+	{
+		.cmd = SMC_NETLINK_DISABLE_SEID,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_disable_seid,
+	},
+	{
+		.cmd = SMC_NETLINK_DUMP_HS_LIMITATION,
+		/* can be retrieved by unprivileged users */
+		.dumpit = smc_nl_dump_hs_limitation,
+	},
+	{
+		.cmd = SMC_NETLINK_ENABLE_HS_LIMITATION,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_enable_hs_limitation,
+	},
+	{
+		.cmd = SMC_NETLINK_DISABLE_HS_LIMITATION,
+		.flags = GENL_ADMIN_PERM,
+		.doit = smc_nl_disable_hs_limitation,
+	},
+};
+
+static const struct nla_policy smc_gen_nl_policy[2] = {
+	[SMC_CMD_MAX_ATTR]	= { .type = NLA_REJECT, },
+};
+
+/* SMC_GENL family definition */
+struct genl_family smc_gen_nl_family __ro_after_init = {
+	.hdrsize =	0,
+	.name =		SMC_GENL_FAMILY_NAME,
+	.version =	SMC_GENL_FAMILY_VERSION,
+	.maxattr =	SMC_CMD_MAX_ATTR,
+	.policy =	smc_gen_nl_policy,
+	.netnsok =	true,
+	.module =	THIS_MODULE,
+	.ops =		smc_gen_nl_ops,
+	.n_ops =	ARRAY_SIZE(smc_gen_nl_ops),
+	.resv_start_op = SMC_NETLINK_DISABLE_HS_LIMITATION + 1,
+};
+
+int __init smc_nl_init(void)
+{
+	return genl_register_family(&smc_gen_nl_family);
+}
+
+void smc_nl_exit(void)
+{
+	genl_unregister_family(&smc_gen_nl_family);
+}

linux-7.2.3/net/smc/smc_netlink.h

@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  SMC Generic netlink operations
+ *
+ *  Copyright IBM Corp. 2020
+ *
+ *  Author(s):	Guvenc Gulce <guvenc@linux.ibm.com>
+ */
+
+#ifndef _SMC_NETLINK_H
+#define _SMC_NETLINK_H
+
+#include <net/netlink.h>
+#include <net/genetlink.h>
+
+extern struct genl_family smc_gen_nl_family;
+
+extern const struct nla_policy smc_gen_ueid_policy[];
+
+struct smc_nl_dmp_ctx {
+	int pos[3];
+};
+
+static inline struct smc_nl_dmp_ctx *smc_nl_dmp_ctx(struct netlink_callback *c)
+{
+	return (struct smc_nl_dmp_ctx *)c->ctx;
+}
+
+int smc_nl_init(void) __init;
+void smc_nl_exit(void);
+
+#endif

linux-7.2.3/net/smc/smc_netns.h

@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Shared Memory Communications
+ *
+ * Network namespace definitions.
+ *
+ * Copyright IBM Corp. 2018
+ */
+
+#ifndef SMC_NETNS_H
+#define SMC_NETNS_H
+
+#include "smc_pnet.h"
+
+extern unsigned int smc_net_id;
+
+/* per-network namespace private data */
+struct smc_net {
+	struct smc_pnettable pnettable;
+	struct smc_pnetids_ndev pnetids_ndev;
+};
+#endif

linux-7.2.3/net/smc/smc_pnet.h

@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  PNET table queries
+ *
+ *  Copyright IBM Corp. 2016
+ *
+ *  Author(s):  Thomas Richter <tmricht@linux.vnet.ibm.com>
+ */
+
+#ifndef _SMC_PNET_H
+#define _SMC_PNET_H
+
+#include <net/smc.h>
+
+#if IS_ENABLED(CONFIG_HAVE_PNETID)
+#include <asm/pnet.h>
+#endif
+
+struct smc_ib_device;
+struct smcd_dev;
+struct smc_init_info;
+struct smc_link_group;
+
+/**
+ * struct smc_pnettable - SMC PNET table anchor
+ * @lock: Lock for list action
+ * @pnetlist: List of PNETIDs
+ */
+struct smc_pnettable {
+	struct mutex lock;
+	struct list_head pnetlist;
+};
+
+struct smc_pnetids_ndev {	/* list of pnetids for net devices in UP state*/
+	struct list_head	list;
+	rwlock_t		lock;
+};
+
+struct smc_pnetids_ndev_entry {
+	struct list_head	list;
+	u8			pnetid[SMC_MAX_PNETID_LEN];
+	refcount_t		refcnt;
+};
+
+static inline int smc_pnetid_by_dev_port(struct device *dev,
+					 unsigned short port, u8 *pnetid)
+{
+#if IS_ENABLED(CONFIG_HAVE_PNETID)
+	return pnet_id_by_dev_port(dev, port, pnetid);
+#else
+	return -ENOENT;
+#endif
+}
+
+int smc_pnet_init(void) __init;
+int smc_pnet_net_init(struct net *net);
+void smc_pnet_exit(void);
+void smc_pnet_net_exit(struct net *net);
+void smc_pnet_find_roce_resource(struct sock *sk, struct smc_init_info *ini);
+void smc_pnet_find_ism_resource(struct sock *sk, struct smc_init_info *ini);
+int smc_pnetid_by_table_ib(struct smc_ib_device *smcibdev, u8 ib_port);
+int smc_pnetid_by_table_smcd(struct smcd_dev *smcd);
+void smc_pnet_find_alt_roce(struct smc_link_group *lgr,
+			    struct smc_init_info *ini,
+			    struct smc_ib_device *known_dev);
+bool smc_pnet_is_ndev_pnetid(struct net *net, u8 *pnetid);
+bool smc_pnet_is_pnetid_set(u8 *pnetid);
+#endif

linux-7.2.3/net/smc/smc_rx.c

@@ -0,0 +1,528 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Manage RMBE
+ * copy new RMBE data into user space
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#include <linux/net.h>
+#include <linux/rcupdate.h>
+#include <linux/sched/signal.h>
+#include <linux/splice.h>
+
+#include <net/sock.h>
+#include <trace/events/sock.h>
+
+#include "smc.h"
+#include "smc_core.h"
+#include "smc_cdc.h"
+#include "smc_tx.h" /* smc_tx_consumer_update() */
+#include "smc_rx.h"
+#include "smc_stats.h"
+#include "smc_tracepoint.h"
+
+/* callback implementation to wakeup consumers blocked with smc_rx_wait().
+ * indirectly called by smc_cdc_msg_recv_action().
+ */
+static void smc_rx_wake_up(struct sock *sk)
+{
+	struct socket_wq *wq;
+
+	trace_sk_data_ready(sk);
+
+	/* derived from sock_def_readable() */
+	/* called already in smc_listen_work() */
+	rcu_read_lock();
+	wq = rcu_dereference(sk->sk_wq);
+	if (skwq_has_sleeper(wq))
+		wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN | EPOLLPRI |
+						EPOLLRDNORM | EPOLLRDBAND);
+	sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_IN);
+	if ((sk->sk_shutdown == SHUTDOWN_MASK) ||
+	    (sk->sk_state == SMC_CLOSED))
+		sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_HUP);
+	rcu_read_unlock();
+}
+
+/* Update consumer cursor
+ *   @conn   connection to update
+ *   @cons   consumer cursor
+ *   @len    number of Bytes consumed
+ *   Returns:
+ *   1 if we should end our receive, 0 otherwise
+ */
+static int smc_rx_update_consumer(struct smc_sock *smc,
+				  union smc_host_cursor cons, size_t len)
+{
+	struct smc_connection *conn = &smc->conn;
+	struct sock *sk = &smc->sk;
+	bool force = false;
+	int diff, rc = 0;
+
+	smc_curs_add(conn->rmb_desc->len, &cons, len);
+
+	/* did we process urgent data? */
+	if (conn->urg_state == SMC_URG_VALID || conn->urg_rx_skip_pend) {
+		diff = smc_curs_comp(conn->rmb_desc->len, &cons,
+				     &conn->urg_curs);
+		if (sock_flag(sk, SOCK_URGINLINE)) {
+			if (diff == 0) {
+				force = true;
+				rc = 1;
+				conn->urg_state = SMC_URG_READ;
+			}
+		} else {
+			if (diff == 1) {
+				/* skip urgent byte */
+				force = true;
+				smc_curs_add(conn->rmb_desc->len, &cons, 1);
+				conn->urg_rx_skip_pend = false;
+			} else if (diff < -1)
+				/* we read past urgent byte */
+				conn->urg_state = SMC_URG_READ;
+		}
+	}
+
+	smc_curs_copy(&conn->local_tx_ctrl.cons, &cons, conn);
+
+	/* send consumer cursor update if required */
+	/* similar to advertising new TCP rcv_wnd if required */
+	smc_tx_consumer_update(conn, force);
+
+	return rc;
+}
+
+static void smc_rx_update_cons(struct smc_sock *smc, size_t len)
+{
+	struct smc_connection *conn = &smc->conn;
+	union smc_host_cursor cons;
+
+	smc_curs_copy(&cons, &conn->local_tx_ctrl.cons, conn);
+	smc_rx_update_consumer(smc, cons, len);
+}
+
+struct smc_spd_priv {
+	struct smc_sock *smc;
+	size_t		 len;
+};
+
+static void smc_rx_pipe_buf_release(struct pipe_inode_info *pipe,
+				    struct pipe_buffer *buf)
+{
+	struct smc_spd_priv *priv = (struct smc_spd_priv *)buf->private;
+	struct smc_sock *smc = priv->smc;
+	struct smc_connection *conn;
+	struct sock *sk = &smc->sk;
+
+	if (sk->sk_state == SMC_CLOSED ||
+	    sk->sk_state == SMC_PEERFINCLOSEWAIT ||
+	    sk->sk_state == SMC_APPFINCLOSEWAIT)
+		goto out;
+	conn = &smc->conn;
+	lock_sock(sk);
+	smc_rx_update_cons(smc, priv->len);
+	release_sock(sk);
+	if (atomic_sub_and_test(priv->len, &conn->splice_pending))
+		smc_rx_wake_up(sk);
+out:
+	kfree(priv);
+	put_page(buf->page);
+	sock_put(sk);
+}
+
+static bool smc_rx_pipe_buf_get(struct pipe_inode_info *pipe,
+				struct pipe_buffer *buf)
+{
+	/* smc_spd_priv in buf->private is not shareable; disallow cloning. */
+	return false;
+}
+
+static const struct pipe_buf_operations smc_pipe_ops = {
+	.release = smc_rx_pipe_buf_release,
+	.get	 = smc_rx_pipe_buf_get,
+};
+
+static void smc_rx_spd_release(struct splice_pipe_desc *spd,
+			       unsigned int i)
+{
+	struct smc_spd_priv *priv = (struct smc_spd_priv *)spd->partial[i].private;
+	struct sock *sk = &priv->smc->sk;
+
+	kfree(priv);
+	put_page(spd->pages[i]);
+	sock_put(sk);
+}
+
+static int smc_rx_splice(struct pipe_inode_info *pipe, char *src, size_t len,
+			 struct smc_sock *smc)
+{
+	struct smc_link_group *lgr = smc->conn.lgr;
+	int offset = offset_in_page(src);
+	struct partial_page *partial;
+	struct splice_pipe_desc spd;
+	struct smc_spd_priv **priv;
+	struct page **pages;
+	int bytes, nr_pages;
+	int i;
+
+	nr_pages = !lgr->is_smcd && smc->conn.rmb_desc->is_vm ?
+		   PAGE_ALIGN(len + offset) / PAGE_SIZE : 1;
+
+	pages = kzalloc_objs(*pages, nr_pages);
+	if (!pages)
+		goto out;
+	partial = kzalloc_objs(*partial, nr_pages);
+	if (!partial)
+		goto out_page;
+	priv = kzalloc_objs(*priv, nr_pages);
+	if (!priv)
+		goto out_part;
+	for (i = 0; i < nr_pages; i++) {
+		priv[i] = kzalloc_obj(**priv);
+		if (!priv[i])
+			goto out_priv;
+	}
+
+	if (lgr->is_smcd ||
+	    (!lgr->is_smcd && !smc->conn.rmb_desc->is_vm)) {
+		/* smcd or smcr that uses physically contiguous RMBs */
+		priv[0]->len = len;
+		priv[0]->smc = smc;
+		partial[0].offset = src - (char *)smc->conn.rmb_desc->cpu_addr;
+		partial[0].len = len;
+		partial[0].private = (unsigned long)priv[0];
+		pages[0] = smc->conn.rmb_desc->pages;
+	} else {
+		int size, left = len;
+		void *buf = src;
+		/* smcr that uses virtually contiguous RMBs*/
+		for (i = 0; i < nr_pages; i++) {
+			size = min_t(int, PAGE_SIZE - offset, left);
+			priv[i]->len = size;
+			priv[i]->smc = smc;
+			pages[i] = vmalloc_to_page(buf);
+			partial[i].offset = offset;
+			partial[i].len = size;
+			partial[i].private = (unsigned long)priv[i];
+			buf += size;
+			left -= size;
+			offset = 0;
+		}
+	}
+	for (i = 0; i < nr_pages; i++) {
+		get_page(pages[i]);
+		sock_hold(&smc->sk);
+	}
+	spd.nr_pages_max = nr_pages;
+	spd.nr_pages = nr_pages;
+	spd.pages = pages;
+	spd.partial = partial;
+	spd.ops = &smc_pipe_ops;
+	spd.spd_release = smc_rx_spd_release;
+
+	bytes = splice_to_pipe(pipe, &spd);
+	if (bytes > 0)
+		atomic_add(bytes, &smc->conn.splice_pending);
+	kfree(priv);
+	kfree(partial);
+	kfree(pages);
+
+	return bytes;
+
+out_priv:
+	for (i = (i - 1); i >= 0; i--)
+		kfree(priv[i]);
+	kfree(priv);
+out_part:
+	kfree(partial);
+out_page:
+	kfree(pages);
+out:
+	return -ENOMEM;
+}
+
+static int smc_rx_data_available_and_no_splice_pend(struct smc_connection *conn, size_t peeked)
+{
+	return smc_rx_data_available(conn, peeked) &&
+	       !atomic_read(&conn->splice_pending);
+}
+
+/* blocks rcvbuf consumer until >=len bytes available or timeout or interrupted
+ *   @smc    smc socket
+ *   @timeo  pointer to max seconds to wait, pointer to value 0 for no timeout
+ *   @peeked  number of bytes already peeked
+ *   @fcrit  add'l criterion to evaluate as function pointer
+ * Returns:
+ * 1 if at least 1 byte available in rcvbuf or if socket error/shutdown.
+ * 0 otherwise (nothing in rcvbuf nor timeout, e.g. interrupted).
+ */
+int smc_rx_wait(struct smc_sock *smc, long *timeo, size_t peeked,
+		int (*fcrit)(struct smc_connection *conn, size_t baseline))
+{
+	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	struct smc_connection *conn = &smc->conn;
+	struct smc_cdc_conn_state_flags *cflags =
+					&conn->local_tx_ctrl.conn_state_flags;
+	struct sock *sk = &smc->sk;
+	int rc;
+
+	if (fcrit(conn, peeked))
+		return 1;
+	sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk);
+	add_wait_queue(sk_sleep(sk), &wait);
+	rc = sk_wait_event(sk, timeo,
+			   READ_ONCE(sk->sk_err) ||
+			   cflags->peer_conn_abort ||
+			   READ_ONCE(sk->sk_shutdown) & RCV_SHUTDOWN ||
+			   conn->killed ||
+			   fcrit(conn, peeked),
+			   &wait);
+	remove_wait_queue(sk_sleep(sk), &wait);
+	sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk);
+	return rc;
+}
+
+static int smc_rx_recv_urg(struct smc_sock *smc, struct msghdr *msg, int len,
+			   int flags)
+{
+	struct smc_connection *conn = &smc->conn;
+	union smc_host_cursor cons;
+	struct sock *sk = &smc->sk;
+	int rc = 0;
+
+	if (sock_flag(sk, SOCK_URGINLINE) ||
+	    !(conn->urg_state == SMC_URG_VALID) ||
+	    conn->urg_state == SMC_URG_READ)
+		return -EINVAL;
+
+	SMC_STAT_INC(smc, urg_data_cnt);
+	if (conn->urg_state == SMC_URG_VALID) {
+		if (!(flags & MSG_PEEK))
+			smc->conn.urg_state = SMC_URG_READ;
+		msg->msg_flags |= MSG_OOB;
+		if (len > 0) {
+			if (!(flags & MSG_TRUNC))
+				rc = memcpy_to_msg(msg, &conn->urg_rx_byte, 1);
+			len = 1;
+			smc_curs_copy(&cons, &conn->local_tx_ctrl.cons, conn);
+			if (smc_curs_diff(conn->rmb_desc->len, &cons,
+					  &conn->urg_curs) > 1)
+				conn->urg_rx_skip_pend = true;
+			/* Urgent Byte was already accounted for, but trigger
+			 * skipping the urgent byte in non-inline case
+			 */
+			if (!(flags & MSG_PEEK))
+				smc_rx_update_consumer(smc, cons, 0);
+		} else {
+			msg->msg_flags |= MSG_TRUNC;
+		}
+
+		return rc ? -EFAULT : len;
+	}
+
+	if (sk->sk_state == SMC_CLOSED || sk->sk_shutdown & RCV_SHUTDOWN)
+		return 0;
+
+	return -EAGAIN;
+}
+
+static bool smc_rx_recvmsg_data_available(struct smc_sock *smc, size_t peeked)
+{
+	struct smc_connection *conn = &smc->conn;
+
+	if (smc_rx_data_available(conn, peeked))
+		return true;
+	else if (conn->urg_state == SMC_URG_VALID)
+		/* we received a single urgent Byte - skip */
+		smc_rx_update_cons(smc, 0);
+	return false;
+}
+
+/* smc_rx_recvmsg - receive data from RMBE
+ * @msg:	copy data to receive buffer
+ * @pipe:	copy data to pipe if set - indicates splice() call
+ *
+ * rcvbuf consumer: main API called by socket layer.
+ * Called under sk lock.
+ */
+int smc_rx_recvmsg(struct smc_sock *smc, struct msghdr *msg,
+		   struct pipe_inode_info *pipe, size_t len, int flags)
+{
+	size_t copylen, read_done = 0, read_remaining = len, peeked_bytes = 0;
+	size_t chunk_len, chunk_off, chunk_len_sum;
+	struct smc_connection *conn = &smc->conn;
+	int (*func)(struct smc_connection *conn, size_t baseline);
+	union smc_host_cursor cons;
+	int readable, chunk;
+	char *rcvbuf_base;
+	struct sock *sk;
+	int splbytes;
+	long timeo;
+	int target;		/* Read at least these many bytes */
+	int rc;
+
+	if (unlikely(flags & MSG_ERRQUEUE))
+		return -EINVAL; /* future work for sk.sk_family == AF_SMC */
+
+	sk = &smc->sk;
+	if (sk->sk_state == SMC_LISTEN)
+		return -ENOTCONN;
+	if (flags & MSG_OOB)
+		return smc_rx_recv_urg(smc, msg, len, flags);
+	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(smc, !conn->lnk);
+
+	if (len < readable)
+		SMC_STAT_RMB_RX_SIZE_SMALL(smc, !conn->lnk);
+	/* we currently use 1 RMBE per RMB, so RMBE == RMB base addr */
+	rcvbuf_base = conn->rx_off + conn->rmb_desc->cpu_addr;
+
+	do { /* while (read_remaining) */
+		if (read_done >= target || (pipe && read_done))
+			break;
+
+		if (conn->killed)
+			break;
+
+		if (smc_rx_recvmsg_data_available(smc, peeked_bytes))
+			goto copy;
+
+		if (sk->sk_shutdown & RCV_SHUTDOWN) {
+			/* smc_cdc_msg_recv_action() could have run after
+			 * above smc_rx_recvmsg_data_available()
+			 */
+			if (smc_rx_recvmsg_data_available(smc, peeked_bytes))
+				goto copy;
+			break;
+		}
+
+		if (read_done) {
+			if (sk->sk_err ||
+			    sk->sk_state == SMC_CLOSED ||
+			    !timeo ||
+			    signal_pending(current))
+				break;
+		} else {
+			if (sk->sk_err) {
+				read_done = sock_error(sk);
+				break;
+			}
+			if (sk->sk_state == SMC_CLOSED) {
+				if (!sock_flag(sk, SOCK_DONE)) {
+					/* This occurs when user tries to read
+					 * from never connected socket.
+					 */
+					read_done = -ENOTCONN;
+					break;
+				}
+				break;
+			}
+			if (!timeo)
+				return -EAGAIN;
+			if (signal_pending(current)) {
+				read_done = sock_intr_errno(timeo);
+				break;
+			}
+		}
+
+		if (!smc_rx_data_available(conn, peeked_bytes)) {
+			smc_rx_wait(smc, &timeo, peeked_bytes, smc_rx_data_available);
+			continue;
+		}
+
+copy:
+		/* initialize variables for 1st iteration of subsequent loop */
+		/* could be just 1 byte, even after waiting on data above */
+		readable = smc_rx_data_available(conn, peeked_bytes);
+		splbytes = atomic_read(&conn->splice_pending);
+		if (!readable || (msg && splbytes)) {
+			if (splbytes)
+				func = smc_rx_data_available_and_no_splice_pend;
+			else
+				func = smc_rx_data_available;
+			smc_rx_wait(smc, &timeo, peeked_bytes, func);
+			continue;
+		}
+
+		smc_curs_copy(&cons, &conn->local_tx_ctrl.cons, conn);
+		if ((flags & MSG_PEEK) && peeked_bytes)
+			smc_curs_add(conn->rmb_desc->len, &cons, peeked_bytes);
+		/* subsequent splice() calls pick up where previous left */
+		if (splbytes)
+			smc_curs_add(conn->rmb_desc->len, &cons, splbytes);
+		if (conn->urg_state == SMC_URG_VALID &&
+		    sock_flag(&smc->sk, SOCK_URGINLINE) &&
+		    readable > 1)
+			readable--;	/* always stop at urgent Byte */
+		/* not more than what user space asked for */
+		copylen = min_t(size_t, read_remaining, readable);
+		/* determine chunks where to read from rcvbuf */
+		/* either unwrapped case, or 1st chunk of wrapped case */
+		chunk_len = min_t(size_t, copylen, conn->rmb_desc->len -
+				  cons.count);
+		chunk_len_sum = chunk_len;
+		chunk_off = cons.count;
+		smc_rmb_sync_sg_for_cpu(conn);
+		for (chunk = 0; chunk < 2; chunk++) {
+			if (!(flags & MSG_TRUNC)) {
+				if (msg) {
+					rc = memcpy_to_msg(msg, rcvbuf_base +
+							   chunk_off,
+							   chunk_len);
+				} else {
+					rc = smc_rx_splice(pipe, rcvbuf_base +
+							chunk_off, chunk_len,
+							smc);
+				}
+				if (rc < 0) {
+					if (!read_done)
+						read_done = -EFAULT;
+					goto out;
+				}
+			}
+			read_remaining -= chunk_len;
+			read_done += chunk_len;
+			if (flags & MSG_PEEK)
+				peeked_bytes += chunk_len;
+
+			if (chunk_len_sum == copylen)
+				break; /* either on 1st or 2nd iteration */
+			/* prepare next (== 2nd) iteration */
+			chunk_len = copylen - chunk_len; /* remainder */
+			chunk_len_sum += chunk_len;
+			chunk_off = 0; /* modulo offset in recv ring buffer */
+		}
+
+		/* update cursors */
+		if (!(flags & MSG_PEEK)) {
+			/* increased in recv tasklet smc_cdc_msg_rcv() */
+			smp_mb__before_atomic();
+			atomic_sub(copylen, &conn->bytes_to_rcv);
+			/* guarantee 0 <= bytes_to_rcv <= rmb_desc->len */
+			smp_mb__after_atomic();
+			if (msg && smc_rx_update_consumer(smc, cons, copylen))
+				goto out;
+		}
+
+		trace_smc_rx_recvmsg(smc, copylen);
+	} while (read_remaining);
+out:
+	return read_done;
+}
+
+/* Initialize receive properties on connection establishment. NB: not __init! */
+void smc_rx_init(struct smc_sock *smc)
+{
+	smc->sk.sk_data_ready = smc_rx_wake_up;
+	atomic_set(&smc->conn.splice_pending, 0);
+	smc->conn.urg_state = SMC_URG_READ;
+}

linux-7.2.3/net/smc/smc_rx.h

@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Manage RMBE
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef SMC_RX_H
+#define SMC_RX_H
+
+#include <linux/socket.h>
+#include <linux/types.h>
+
+#include "smc.h"
+
+void smc_rx_init(struct smc_sock *smc);
+
+int smc_rx_recvmsg(struct smc_sock *smc, struct msghdr *msg,
+		   struct pipe_inode_info *pipe, size_t len, int flags);
+int smc_rx_wait(struct smc_sock *smc, long *timeo, size_t peeked,
+		int (*fcrit)(struct smc_connection *conn, size_t baseline));
+static inline int smc_rx_data_available(struct smc_connection *conn, size_t peeked)
+{
+	return atomic_read(&conn->bytes_to_rcv) - peeked;
+}
+
+#endif /* SMC_RX_H */

linux-7.2.3/net/smc/smc_stats.c

@@ -0,0 +1,419 @@
+// 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 <linux/smc.h>
+#include <net/genetlink.h>
+#include <net/sock.h>
+#include "smc_netlink.h"
+#include "smc_stats.h"
+
+int smc_stats_init(struct net *net)
+{
+	net->smc.fback_rsn = kzalloc_obj(*net->smc.fback_rsn);
+	if (!net->smc.fback_rsn)
+		goto err_fback;
+	net->smc.smc_stats = alloc_percpu(struct smc_stats);
+	if (!net->smc.smc_stats)
+		goto err_stats;
+	mutex_init(&net->smc.mutex_fback_rsn);
+	return 0;
+
+err_stats:
+	kfree(net->smc.fback_rsn);
+err_fback:
+	return -ENOMEM;
+}
+
+void smc_stats_exit(struct net *net)
+{
+	kfree(net->smc.fback_rsn);
+	if (net->smc.smc_stats)
+		free_percpu(net->smc.smc_stats);
+}
+
+static int smc_nl_fill_stats_rmb_data(struct sk_buff *skb,
+				      struct smc_stats *stats, int tech,
+				      int type)
+{
+	struct smc_stats_rmbcnt *stats_rmb_cnt;
+	struct nlattr *attrs;
+
+	if (type == SMC_NLA_STATS_T_TX_RMB_STATS)
+		stats_rmb_cnt = &stats->smc[tech].rmb_tx;
+	else
+		stats_rmb_cnt = &stats->smc[tech].rmb_rx;
+
+	attrs = nla_nest_start(skb, type);
+	if (!attrs)
+		goto errout;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_REUSE_CNT,
+			      stats_rmb_cnt->reuse_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_SIZE_SM_PEER_CNT,
+			      stats_rmb_cnt->buf_size_small_peer_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_SIZE_SM_CNT,
+			      stats_rmb_cnt->buf_size_small_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_FULL_PEER_CNT,
+			      stats_rmb_cnt->buf_full_peer_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_FULL_CNT,
+			      stats_rmb_cnt->buf_full_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_ALLOC_CNT,
+			      stats_rmb_cnt->alloc_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_RMB_DGRADE_CNT,
+			      stats_rmb_cnt->dgrade_cnt,
+			      SMC_NLA_STATS_RMB_PAD))
+		goto errattr;
+
+	nla_nest_end(skb, attrs);
+	return 0;
+
+errattr:
+	nla_nest_cancel(skb, attrs);
+errout:
+	return -EMSGSIZE;
+}
+
+static int smc_nl_fill_stats_bufsize_data(struct sk_buff *skb,
+					  struct smc_stats *stats, int tech,
+					  int type)
+{
+	struct smc_stats_memsize *stats_pload;
+	struct nlattr *attrs;
+
+	if (type == SMC_NLA_STATS_T_TXPLOAD_SIZE)
+		stats_pload = &stats->smc[tech].tx_pd;
+	else if (type == SMC_NLA_STATS_T_RXPLOAD_SIZE)
+		stats_pload = &stats->smc[tech].rx_pd;
+	else if (type == SMC_NLA_STATS_T_TX_RMB_SIZE)
+		stats_pload = &stats->smc[tech].tx_rmbsize;
+	else if (type == SMC_NLA_STATS_T_RX_RMB_SIZE)
+		stats_pload = &stats->smc[tech].rx_rmbsize;
+	else
+		goto errout;
+
+	attrs = nla_nest_start(skb, type);
+	if (!attrs)
+		goto errout;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_8K,
+			      stats_pload->buf[SMC_BUF_8K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_16K,
+			      stats_pload->buf[SMC_BUF_16K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_32K,
+			      stats_pload->buf[SMC_BUF_32K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_64K,
+			      stats_pload->buf[SMC_BUF_64K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_128K,
+			      stats_pload->buf[SMC_BUF_128K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_256K,
+			      stats_pload->buf[SMC_BUF_256K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_512K,
+			      stats_pload->buf[SMC_BUF_512K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_1024K,
+			      stats_pload->buf[SMC_BUF_1024K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_PLOAD_G_1024K,
+			      stats_pload->buf[SMC_BUF_G_1024K],
+			      SMC_NLA_STATS_PLOAD_PAD))
+		goto errattr;
+
+	nla_nest_end(skb, attrs);
+	return 0;
+
+errattr:
+	nla_nest_cancel(skb, attrs);
+errout:
+	return -EMSGSIZE;
+}
+
+static int smc_nl_fill_stats_tech_data(struct sk_buff *skb,
+				       struct smc_stats *stats, int tech)
+{
+	struct smc_stats_tech *smc_tech;
+	struct nlattr *attrs;
+
+	smc_tech = &stats->smc[tech];
+	if (tech == SMC_TYPE_D)
+		attrs = nla_nest_start(skb, SMC_NLA_STATS_SMCD_TECH);
+	else
+		attrs = nla_nest_start(skb, SMC_NLA_STATS_SMCR_TECH);
+
+	if (!attrs)
+		goto errout;
+	if (smc_nl_fill_stats_rmb_data(skb, stats, tech,
+				       SMC_NLA_STATS_T_TX_RMB_STATS))
+		goto errattr;
+	if (smc_nl_fill_stats_rmb_data(skb, stats, tech,
+				       SMC_NLA_STATS_T_RX_RMB_STATS))
+		goto errattr;
+	if (smc_nl_fill_stats_bufsize_data(skb, stats, tech,
+					   SMC_NLA_STATS_T_TXPLOAD_SIZE))
+		goto errattr;
+	if (smc_nl_fill_stats_bufsize_data(skb, stats, tech,
+					   SMC_NLA_STATS_T_RXPLOAD_SIZE))
+		goto errattr;
+	if (smc_nl_fill_stats_bufsize_data(skb, stats, tech,
+					   SMC_NLA_STATS_T_TX_RMB_SIZE))
+		goto errattr;
+	if (smc_nl_fill_stats_bufsize_data(skb, stats, tech,
+					   SMC_NLA_STATS_T_RX_RMB_SIZE))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_CLNT_V1_SUCC,
+			      smc_tech->clnt_v1_succ_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_CLNT_V2_SUCC,
+			      smc_tech->clnt_v2_succ_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_SRV_V1_SUCC,
+			      smc_tech->srv_v1_succ_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_SRV_V2_SUCC,
+			      smc_tech->srv_v2_succ_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_RX_BYTES,
+			      smc_tech->rx_bytes,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_TX_BYTES,
+			      smc_tech->tx_bytes,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_uint(skb, SMC_NLA_STATS_T_RX_RMB_USAGE,
+			 smc_tech->rx_rmbuse))
+		goto errattr;
+	if (nla_put_uint(skb, SMC_NLA_STATS_T_TX_RMB_USAGE,
+			 smc_tech->tx_rmbuse))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_RX_CNT,
+			      smc_tech->rx_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_TX_CNT,
+			      smc_tech->tx_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_SENDPAGE_CNT,
+			      0,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_CORK_CNT,
+			      smc_tech->cork_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_NDLY_CNT,
+			      smc_tech->ndly_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_SPLICE_CNT,
+			      smc_tech->splice_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_T_URG_DATA_CNT,
+			      smc_tech->urg_data_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+
+	nla_nest_end(skb, attrs);
+	return 0;
+
+errattr:
+	nla_nest_cancel(skb, attrs);
+errout:
+	return -EMSGSIZE;
+}
+
+int smc_nl_get_stats(struct sk_buff *skb,
+		     struct netlink_callback *cb)
+{
+	struct smc_nl_dmp_ctx *cb_ctx = smc_nl_dmp_ctx(cb);
+	struct net *net = sock_net(skb->sk);
+	struct smc_stats *stats;
+	struct nlattr *attrs;
+	int cpu, i, size;
+	void *nlh;
+	u64 *src;
+	u64 *sum;
+
+	if (cb_ctx->pos[0])
+		goto errmsg;
+	nlh = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			  &smc_gen_nl_family, NLM_F_MULTI,
+			  SMC_NETLINK_GET_STATS);
+	if (!nlh)
+		goto errmsg;
+
+	attrs = nla_nest_start(skb, SMC_GEN_STATS);
+	if (!attrs)
+		goto errnest;
+	stats = kzalloc_obj(*stats);
+	if (!stats)
+		goto erralloc;
+	size = sizeof(*stats) / sizeof(u64);
+	for_each_possible_cpu(cpu) {
+		src = (u64 *)per_cpu_ptr(net->smc.smc_stats, cpu);
+		sum = (u64 *)stats;
+		for (i = 0; i < size; i++)
+			*(sum++) += *(src++);
+	}
+	if (smc_nl_fill_stats_tech_data(skb, stats, SMC_TYPE_D))
+		goto errattr;
+	if (smc_nl_fill_stats_tech_data(skb, stats, SMC_TYPE_R))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_CLNT_HS_ERR_CNT,
+			      stats->clnt_hshake_err_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+	if (nla_put_u64_64bit(skb, SMC_NLA_STATS_SRV_HS_ERR_CNT,
+			      stats->srv_hshake_err_cnt,
+			      SMC_NLA_STATS_PAD))
+		goto errattr;
+
+	nla_nest_end(skb, attrs);
+	genlmsg_end(skb, nlh);
+	cb_ctx->pos[0] = 1;
+	kfree(stats);
+	return skb->len;
+
+errattr:
+	kfree(stats);
+erralloc:
+	nla_nest_cancel(skb, attrs);
+errnest:
+	genlmsg_cancel(skb, nlh);
+errmsg:
+	return skb->len;
+}
+
+static int smc_nl_get_fback_details(struct sk_buff *skb,
+				    struct netlink_callback *cb, int pos,
+				    bool is_srv)
+{
+	struct smc_nl_dmp_ctx *cb_ctx = smc_nl_dmp_ctx(cb);
+	struct net *net = sock_net(skb->sk);
+	int cnt_reported = cb_ctx->pos[2];
+	struct smc_stats_fback *trgt_arr;
+	struct nlattr *attrs;
+	int rc = 0;
+	void *nlh;
+
+	if (is_srv)
+		trgt_arr = &net->smc.fback_rsn->srv[0];
+	else
+		trgt_arr = &net->smc.fback_rsn->clnt[0];
+	if (!trgt_arr[pos].fback_code)
+		return -ENODATA;
+	nlh = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			  &smc_gen_nl_family, NLM_F_MULTI,
+			  SMC_NETLINK_GET_FBACK_STATS);
+	if (!nlh)
+		goto errmsg;
+	attrs = nla_nest_start(skb, SMC_GEN_FBACK_STATS);
+	if (!attrs)
+		goto errout;
+	if (nla_put_u8(skb, SMC_NLA_FBACK_STATS_TYPE, is_srv))
+		goto errattr;
+	if (!cnt_reported) {
+		if (nla_put_u64_64bit(skb, SMC_NLA_FBACK_STATS_SRV_CNT,
+				      net->smc.fback_rsn->srv_fback_cnt,
+				      SMC_NLA_FBACK_STATS_PAD))
+			goto errattr;
+		if (nla_put_u64_64bit(skb, SMC_NLA_FBACK_STATS_CLNT_CNT,
+				      net->smc.fback_rsn->clnt_fback_cnt,
+				      SMC_NLA_FBACK_STATS_PAD))
+			goto errattr;
+		cnt_reported = 1;
+	}
+
+	if (nla_put_u32(skb, SMC_NLA_FBACK_STATS_RSN_CODE,
+			trgt_arr[pos].fback_code))
+		goto errattr;
+	if (nla_put_u16(skb, SMC_NLA_FBACK_STATS_RSN_CNT,
+			trgt_arr[pos].count))
+		goto errattr;
+
+	cb_ctx->pos[2] = cnt_reported;
+	nla_nest_end(skb, attrs);
+	genlmsg_end(skb, nlh);
+	return rc;
+
+errattr:
+	nla_nest_cancel(skb, attrs);
+errout:
+	genlmsg_cancel(skb, nlh);
+errmsg:
+	return -EMSGSIZE;
+}
+
+int smc_nl_get_fback_stats(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct smc_nl_dmp_ctx *cb_ctx = smc_nl_dmp_ctx(cb);
+	struct net *net = sock_net(skb->sk);
+	int rc_srv = 0, rc_clnt = 0, k;
+	int skip_serv = cb_ctx->pos[1];
+	int snum = cb_ctx->pos[0];
+	bool is_srv = true;
+
+	mutex_lock(&net->smc.mutex_fback_rsn);
+	for (k = 0; k < SMC_MAX_FBACK_RSN_CNT; k++) {
+		if (k < snum)
+			continue;
+		if (!skip_serv) {
+			rc_srv = smc_nl_get_fback_details(skb, cb, k, is_srv);
+			if (rc_srv && rc_srv != -ENODATA)
+				break;
+		} else {
+			skip_serv = 0;
+		}
+		rc_clnt = smc_nl_get_fback_details(skb, cb, k, !is_srv);
+		if (rc_clnt && rc_clnt != -ENODATA) {
+			skip_serv = 1;
+			break;
+		}
+		if (rc_clnt == -ENODATA && rc_srv == -ENODATA)
+			break;
+	}
+	mutex_unlock(&net->smc.mutex_fback_rsn);
+	cb_ctx->pos[1] = skip_serv;
+	cb_ctx->pos[0] = k;
+	return skb->len;
+}

linux-7.2.3/net/smc/smc_stats.h

@@ -0,0 +1,280 @@
+/* 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 36
+
+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_rsn {
+	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			urg_data_cnt;
+	u64			splice_cnt;
+	u64			cork_cnt;
+	u64			ndly_cnt;
+	u64			rx_bytes;
+	u64			tx_bytes;
+	u64			rx_cnt;
+	u64			tx_cnt;
+	u64			rx_rmbuse;
+	u64			tx_rmbuse;
+};
+
+struct smc_stats {
+	struct smc_stats_tech	smc[2];
+	u64			clnt_hshake_err_cnt;
+	u64			srv_hshake_err_cnt;
+};
+
+#define SMC_STAT_PAYLOAD_SUB(_smc_stats, _tech, key, _len, _rc) \
+do { \
+	typeof(_smc_stats) stats = (_smc_stats); \
+	typeof(_tech) t = (_tech); \
+	typeof(_len) l = (_len); \
+	int _pos; \
+	typeof(_rc) r = (_rc); \
+	int m = SMC_BUF_MAX - 1; \
+	this_cpu_inc((*stats).smc[t].key ## _cnt); \
+	if (r <= 0 || l <= 0) \
+		break; \
+	_pos = fls64((l - 1) >> 13); \
+	_pos = (_pos <= m) ? _pos : m; \
+	this_cpu_inc((*stats).smc[t].key ## _pd.buf[_pos]); \
+	this_cpu_add((*stats).smc[t].key ## _bytes, r); \
+} \
+while (0)
+
+#define SMC_STAT_TX_PAYLOAD(_smc, length, rcode) \
+do { \
+	typeof(_smc) __smc = _smc; \
+	struct net *_net = sock_net(&__smc->sk); \
+	struct smc_stats __percpu *_smc_stats = _net->smc.smc_stats; \
+	typeof(length) _len = (length); \
+	typeof(rcode) _rc = (rcode); \
+	bool is_smcd = !__smc->conn.lnk; \
+	if (is_smcd) \
+		SMC_STAT_PAYLOAD_SUB(_smc_stats, SMC_TYPE_D, tx, _len, _rc); \
+	else \
+		SMC_STAT_PAYLOAD_SUB(_smc_stats, SMC_TYPE_R, tx, _len, _rc); \
+} \
+while (0)
+
+#define SMC_STAT_RX_PAYLOAD(_smc, length, rcode) \
+do { \
+	typeof(_smc) __smc = _smc; \
+	struct net *_net = sock_net(&__smc->sk); \
+	struct smc_stats __percpu *_smc_stats = _net->smc.smc_stats; \
+	typeof(length) _len = (length); \
+	typeof(rcode) _rc = (rcode); \
+	bool is_smcd = !__smc->conn.lnk; \
+	if (is_smcd) \
+		SMC_STAT_PAYLOAD_SUB(_smc_stats, SMC_TYPE_D, rx, _len, _rc); \
+	else \
+		SMC_STAT_PAYLOAD_SUB(_smc_stats, SMC_TYPE_R, rx, _len, _rc); \
+} \
+while (0)
+
+#define SMC_STAT_RMB_SIZE_SUB(_smc_stats, _tech, k, _is_add, _len) \
+do { \
+	typeof(_smc_stats) stats = (_smc_stats); \
+	typeof(_is_add) is_a = (_is_add); \
+	typeof(_len) _l = (_len); \
+	typeof(_tech) t = (_tech); \
+	int _pos; \
+	int m = SMC_BUF_MAX - 1; \
+	if (_l <= 0) \
+		break; \
+	if (is_a) { \
+		_pos = fls((_l - 1) >> 13); \
+		_pos = (_pos <= m) ? _pos : m; \
+		this_cpu_inc((*stats).smc[t].k ## _rmbsize.buf[_pos]); \
+		this_cpu_add((*stats).smc[t].k ## _rmbuse, _l); \
+	} else { \
+		this_cpu_sub((*stats).smc[t].k ## _rmbuse, _l); \
+	} \
+} \
+while (0)
+
+#define SMC_STAT_RMB_SUB(_smc_stats, type, t, key) \
+	this_cpu_inc((*(_smc_stats)).smc[t].rmb ## _ ## key.type ## _cnt)
+
+#define SMC_STAT_RMB_SIZE(_smc, _is_smcd, _is_rx, _is_add, _len) \
+do { \
+	struct net *_net = sock_net(&(_smc)->sk); \
+	struct smc_stats __percpu *_smc_stats = _net->smc.smc_stats; \
+	typeof(_is_add) is_add = (_is_add); \
+	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_stats, SMC_TYPE_D, rx, is_add, l); \
+	if ((is_d) && !(is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(_smc_stats, SMC_TYPE_D, tx, is_add, l); \
+	if (!(is_d) && (is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(_smc_stats, SMC_TYPE_R, rx, is_add, l); \
+	if (!(is_d) && !(is_r)) \
+		SMC_STAT_RMB_SIZE_SUB(_smc_stats, SMC_TYPE_R, tx, is_add, l); \
+} \
+while (0)
+
+#define SMC_STAT_RMB(_smc, type, _is_smcd, _is_rx) \
+do { \
+	struct net *net = sock_net(&(_smc)->sk); \
+	struct smc_stats __percpu *_smc_stats = net->smc.smc_stats; \
+	typeof(_is_smcd) is_d = (_is_smcd); \
+	typeof(_is_rx) is_r = (_is_rx); \
+	if ((is_d) && (is_r)) \
+		SMC_STAT_RMB_SUB(_smc_stats, type, SMC_TYPE_D, rx); \
+	if ((is_d) && !(is_r)) \
+		SMC_STAT_RMB_SUB(_smc_stats, type, SMC_TYPE_D, tx); \
+	if (!(is_d) && (is_r)) \
+		SMC_STAT_RMB_SUB(_smc_stats, type, SMC_TYPE_R, rx); \
+	if (!(is_d) && !(is_r)) \
+		SMC_STAT_RMB_SUB(_smc_stats, type, SMC_TYPE_R, tx); \
+} \
+while (0)
+
+#define SMC_STAT_BUF_REUSE(smc, is_smcd, is_rx) \
+	SMC_STAT_RMB(smc, reuse, is_smcd, is_rx)
+
+#define SMC_STAT_RMB_ALLOC(smc, is_smcd, is_rx) \
+	SMC_STAT_RMB(smc, alloc, is_smcd, is_rx)
+
+#define SMC_STAT_RMB_DOWNGRADED(smc, is_smcd, is_rx) \
+	SMC_STAT_RMB(smc, dgrade, is_smcd, is_rx)
+
+#define SMC_STAT_RMB_TX_PEER_FULL(smc, is_smcd) \
+	SMC_STAT_RMB(smc, buf_full_peer, is_smcd, false)
+
+#define SMC_STAT_RMB_TX_FULL(smc, is_smcd) \
+	SMC_STAT_RMB(smc, buf_full, is_smcd, false)
+
+#define SMC_STAT_RMB_TX_PEER_SIZE_SMALL(smc, is_smcd) \
+	SMC_STAT_RMB(smc, buf_size_small_peer, is_smcd, false)
+
+#define SMC_STAT_RMB_TX_SIZE_SMALL(smc, is_smcd) \
+	SMC_STAT_RMB(smc, buf_size_small, is_smcd, false)
+
+#define SMC_STAT_RMB_RX_SIZE_SMALL(smc, is_smcd) \
+	SMC_STAT_RMB(smc, buf_size_small, is_smcd, true)
+
+#define SMC_STAT_RMB_RX_FULL(smc, is_smcd) \
+	SMC_STAT_RMB(smc, buf_full, is_smcd, true)
+
+#define SMC_STAT_INC(_smc, type) \
+do { \
+	typeof(_smc) __smc = _smc; \
+	bool is_smcd = !(__smc)->conn.lnk; \
+	struct net *net = sock_net(&(__smc)->sk); \
+	struct smc_stats __percpu *smc_stats = net->smc.smc_stats; \
+	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(net, _aclc) \
+do { \
+	typeof(_aclc) acl = (_aclc); \
+	bool is_v2 = (acl->hdr.version == SMC_V2); \
+	bool is_smcd = (acl->hdr.typev1 == SMC_TYPE_D); \
+	struct smc_stats __percpu *smc_stats = (net)->smc.smc_stats; \
+	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(net, _ini) \
+do { \
+	typeof(_ini) i = (_ini); \
+	bool is_smcd = (i->is_smcd); \
+	u8 version = is_smcd ? i->smcd_version : i->smcr_version; \
+	bool is_v2 = (version & SMC_V2); \
+	typeof(net->smc.smc_stats) smc_stats = (net)->smc.smc_stats; \
+	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_nl_get_stats(struct sk_buff *skb, struct netlink_callback *cb);
+int smc_nl_get_fback_stats(struct sk_buff *skb, struct netlink_callback *cb);
+int smc_stats_init(struct net *net);
+void smc_stats_exit(struct net *net);
+
+#endif /* NET_SMC_SMC_STATS_H_ */

linux-7.2.3/net/smc/smc_sysctl.c

@@ -0,0 +1,266 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  smc_sysctl.c: sysctl interface to SMC subsystem.
+ *
+ *  Copyright (c) 2022, Alibaba Inc.
+ *
+ *  Author: Tony Lu <tonylu@linux.alibaba.com>
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/sysctl.h>
+#include <linux/bpf.h>
+#include <net/net_namespace.h>
+
+#include "smc.h"
+#include "smc_core.h"
+#include "smc_llc.h"
+#include "smc_sysctl.h"
+#include "smc_hs_bpf.h"
+
+static int min_sndbuf = SMC_BUF_MIN_SIZE;
+static int min_rcvbuf = SMC_BUF_MIN_SIZE;
+static int max_sndbuf = INT_MAX / 2;
+static int max_rcvbuf = INT_MAX / 2;
+static const int net_smc_wmem_init = (64 * 1024);
+static const int net_smc_rmem_init = (64 * 1024);
+static int links_per_lgr_min = SMC_LINKS_ADD_LNK_MIN;
+static int links_per_lgr_max = SMC_LINKS_ADD_LNK_MAX;
+static int conns_per_lgr_min = SMC_CONN_PER_LGR_MIN;
+static int conns_per_lgr_max = SMC_CONN_PER_LGR_MAX;
+static unsigned int smcr_max_wr_min = 2;
+static unsigned int smcr_max_wr_max = 2048;
+
+#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)
+static int smc_net_replace_smc_hs_ctrl(struct net *net, const char *name)
+{
+	struct smc_hs_ctrl *ctrl = NULL;
+
+	rcu_read_lock();
+	/* null or empty name ask to clear current ctrl */
+	if (name && name[0]) {
+		ctrl = smc_hs_ctrl_find_by_name(name);
+		if (!ctrl) {
+			rcu_read_unlock();
+			return -EINVAL;
+		}
+		/* no change, just return */
+		if (ctrl == rcu_dereference(net->smc.hs_ctrl)) {
+			rcu_read_unlock();
+			return 0;
+		}
+		if (!bpf_try_module_get(ctrl, ctrl->owner)) {
+			rcu_read_unlock();
+			return -EBUSY;
+		}
+	}
+	/* xhcg old ctrl with the new one atomically */
+	ctrl = unrcu_pointer(xchg(&net->smc.hs_ctrl, RCU_INITIALIZER(ctrl)));
+	/* release old ctrl */
+	if (ctrl)
+		bpf_module_put(ctrl, ctrl->owner);
+
+	rcu_read_unlock();
+	return 0;
+}
+
+static int proc_smc_hs_ctrl(const struct ctl_table *ctl, int write,
+			    void *buffer, size_t *lenp, loff_t *ppos)
+{
+	struct net *net = container_of(ctl->data, struct net, smc.hs_ctrl);
+	char val[SMC_HS_CTRL_NAME_MAX];
+	const struct ctl_table tbl = {
+		.data = val,
+		.maxlen = SMC_HS_CTRL_NAME_MAX,
+	};
+	struct smc_hs_ctrl *ctrl;
+	int ret;
+
+	rcu_read_lock();
+	ctrl = rcu_dereference(net->smc.hs_ctrl);
+	if (ctrl)
+		memcpy(val, ctrl->name, sizeof(ctrl->name));
+	else
+		val[0] = '\0';
+	rcu_read_unlock();
+
+	ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
+	if (ret)
+		return ret;
+
+	if (write)
+		ret = smc_net_replace_smc_hs_ctrl(net, val);
+	return ret;
+}
+#endif /* CONFIG_SMC_HS_CTRL_BPF */
+
+static struct ctl_table smc_table[] = {
+	{
+		.procname       = "autocorking_size",
+		.data           = &init_net.smc.sysctl_autocorking_size,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler	= proc_douintvec,
+	},
+	{
+		.procname	= "smcr_buf_type",
+		.data		= &init_net.smc.sysctl_smcr_buf_type,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_douintvec_minmax,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_TWO,
+	},
+	{
+		.procname	= "smcr_testlink_time",
+		.data		= &init_net.smc.sysctl_smcr_testlink_time,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.procname	= "wmem",
+		.data		= &init_net.smc.sysctl_wmem,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_sndbuf,
+		.extra2		= &max_sndbuf,
+	},
+	{
+		.procname	= "rmem",
+		.data		= &init_net.smc.sysctl_rmem,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_rcvbuf,
+		.extra2		= &max_rcvbuf,
+	},
+	{
+		.procname	= "smcr_max_links_per_lgr",
+		.data		= &init_net.smc.sysctl_max_links_per_lgr,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &links_per_lgr_min,
+		.extra2		= &links_per_lgr_max,
+	},
+	{
+		.procname	= "smcr_max_conns_per_lgr",
+		.data		= &init_net.smc.sysctl_max_conns_per_lgr,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &conns_per_lgr_min,
+		.extra2		= &conns_per_lgr_max,
+	},
+	{
+		.procname	= "limit_smc_hs",
+		.data		= &init_net.smc.limit_smc_hs,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_ONE,
+	},
+	{
+		.procname	= "smcr_max_send_wr",
+		.data		= &init_net.smc.sysctl_smcr_max_send_wr,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &smcr_max_wr_min,
+		.extra2		= &smcr_max_wr_max,
+	},
+	{
+		.procname	= "smcr_max_recv_wr",
+		.data		= &init_net.smc.sysctl_smcr_max_recv_wr,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &smcr_max_wr_min,
+		.extra2		= &smcr_max_wr_max,
+	},
+#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)
+	{
+		.procname	= "hs_ctrl",
+		.data		= &init_net.smc.hs_ctrl,
+		.mode		= 0644,
+		.maxlen		= SMC_HS_CTRL_NAME_MAX,
+		.proc_handler	= proc_smc_hs_ctrl,
+	},
+#endif /* CONFIG_SMC_HS_CTRL_BPF */
+};
+
+int __net_init smc_sysctl_net_init(struct net *net)
+{
+	size_t table_size = ARRAY_SIZE(smc_table);
+	struct ctl_table *table;
+
+	table = smc_table;
+	if (!net_eq(net, &init_net)) {
+		int i;
+#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)
+		struct smc_hs_ctrl *ctrl;
+
+		rcu_read_lock();
+		ctrl = rcu_dereference(init_net.smc.hs_ctrl);
+		if (ctrl && ctrl->flags & SMC_HS_CTRL_FLAG_INHERITABLE &&
+		    bpf_try_module_get(ctrl, ctrl->owner))
+			rcu_assign_pointer(net->smc.hs_ctrl, ctrl);
+		rcu_read_unlock();
+#endif /* CONFIG_SMC_HS_CTRL_BPF */
+
+		table = kmemdup(table, sizeof(smc_table), GFP_KERNEL);
+		if (!table)
+			goto err_alloc;
+
+		for (i = 0; i < table_size; i++)
+			table[i].data += (void *)net - (void *)&init_net;
+	}
+
+	net->smc.smc_hdr = register_net_sysctl_sz(net, "net/smc", table,
+						  table_size);
+	if (!net->smc.smc_hdr)
+		goto err_reg;
+
+	net->smc.sysctl_autocorking_size = SMC_AUTOCORKING_DEFAULT_SIZE;
+	net->smc.sysctl_smcr_buf_type = SMCR_PHYS_CONT_BUFS;
+	net->smc.sysctl_smcr_testlink_time = SMC_LLC_TESTLINK_DEFAULT_TIME;
+	WRITE_ONCE(net->smc.sysctl_wmem, net_smc_wmem_init);
+	WRITE_ONCE(net->smc.sysctl_rmem, net_smc_rmem_init);
+	net->smc.sysctl_max_links_per_lgr = SMC_LINKS_PER_LGR_MAX_PREFER;
+	net->smc.sysctl_max_conns_per_lgr = SMC_CONN_PER_LGR_PREFER;
+	net->smc.sysctl_smcr_max_send_wr = SMCR_MAX_SEND_WR_DEF;
+	net->smc.sysctl_smcr_max_recv_wr = SMCR_MAX_RECV_WR_DEF;
+	/* disable handshake limitation by default */
+	net->smc.limit_smc_hs = 0;
+
+	return 0;
+
+err_reg:
+	if (!net_eq(net, &init_net))
+		kfree(table);
+err_alloc:
+#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)
+	smc_net_replace_smc_hs_ctrl(net, NULL);
+#endif /* CONFIG_SMC_HS_CTRL_BPF */
+	return -ENOMEM;
+}
+
+void __net_exit smc_sysctl_net_exit(struct net *net)
+{
+	const struct ctl_table *table;
+
+	table = net->smc.smc_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->smc.smc_hdr);
+#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)
+	smc_net_replace_smc_hs_ctrl(net, NULL);
+#endif /* CONFIG_SMC_HS_CTRL_BPF */
+
+	if (!net_eq(net, &init_net))
+		kfree(table);
+}

linux-7.2.3/net/smc/smc_sysctl.h

@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  smc_sysctl.c: sysctl interface to SMC subsystem.
+ *
+ *  Copyright (c) 2022, Alibaba Inc.
+ *
+ *  Author: Tony Lu <tonylu@linux.alibaba.com>
+ *
+ */
+
+#ifndef _SMC_SYSCTL_H
+#define _SMC_SYSCTL_H
+
+#ifdef CONFIG_SYSCTL
+
+int __net_init smc_sysctl_net_init(struct net *net);
+void __net_exit smc_sysctl_net_exit(struct net *net);
+
+#else
+
+static inline int smc_sysctl_net_init(struct net *net)
+{
+	net->smc.sysctl_autocorking_size = SMC_AUTOCORKING_DEFAULT_SIZE;
+	net->smc.sysctl_max_links_per_lgr = SMC_LINKS_PER_LGR_MAX_PREFER;
+	net->smc.sysctl_max_conns_per_lgr = SMC_CONN_PER_LGR_PREFER;
+	net->smc.sysctl_smcr_max_send_wr = SMCR_MAX_SEND_WR_DEF;
+	net->smc.sysctl_smcr_max_recv_wr = SMCR_MAX_RECV_WR_DEF;
+	return 0;
+}
+
+static inline void smc_sysctl_net_exit(struct net *net) { }
+
+#endif /* CONFIG_SYSCTL */
+
+#endif /* _SMC_SYSCTL_H */

linux-7.2.3/net/smc/smc_tracepoint.c

@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#define CREATE_TRACE_POINTS
+#include "smc_tracepoint.h"
+
+EXPORT_TRACEPOINT_SYMBOL(smc_switch_to_fallback);
+EXPORT_TRACEPOINT_SYMBOL(smc_tx_sendmsg);
+EXPORT_TRACEPOINT_SYMBOL(smc_rx_recvmsg);
+EXPORT_TRACEPOINT_SYMBOL(smcr_link_down);

linux-7.2.3/net/smc/smc_tracepoint.h

@@ -0,0 +1,125 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM smc
+
+#if !defined(_TRACE_SMC_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_SMC_H
+
+#include <linux/ipv6.h>
+#include <linux/tcp.h>
+#include <linux/tracepoint.h>
+#include <net/ipv6.h>
+#include "smc.h"
+#include "smc_core.h"
+
+TRACE_EVENT(smc_switch_to_fallback,
+
+	    TP_PROTO(const struct smc_sock *smc, int fallback_rsn),
+
+	    TP_ARGS(smc, fallback_rsn),
+
+	    TP_STRUCT__entry(
+			     __field(const void *, sk)
+			     __field(const void *, clcsk)
+			     __field(u64, net_cookie)
+			     __field(int, fallback_rsn)
+	    ),
+
+	    TP_fast_assign(
+			   const struct sock *sk = &smc->sk;
+			   const struct sock *clcsk = smc->clcsock->sk;
+
+			   __entry->sk = sk;
+			   __entry->clcsk = clcsk;
+			   __entry->net_cookie = sock_net(sk)->net_cookie;
+			   __entry->fallback_rsn = fallback_rsn;
+	    ),
+
+	    TP_printk("sk=%p clcsk=%p net=%llu fallback_rsn=%d",
+		      __entry->sk, __entry->clcsk,
+		      __entry->net_cookie, __entry->fallback_rsn)
+);
+
+DECLARE_EVENT_CLASS(smc_msg_event,
+
+		    TP_PROTO(const struct smc_sock *smc, size_t len),
+
+		    TP_ARGS(smc, len),
+
+		    TP_STRUCT__entry(
+				     __field(const void *, smc)
+				     __field(u64, net_cookie)
+				     __field(size_t, len)
+				     __string(name, smc->conn.lnk ? smc->conn.lnk->ibname : "")
+		    ),
+
+		    TP_fast_assign(
+				   const struct sock *sk = &smc->sk;
+
+				   __entry->smc = smc;
+				   __entry->net_cookie = sock_net(sk)->net_cookie;
+				   __entry->len = len;
+				   __assign_str(name);
+		    ),
+
+		    TP_printk("smc=%p net=%llu len=%zu dev=%s",
+			      __entry->smc, __entry->net_cookie,
+			      __entry->len, __get_str(name))
+);
+
+DEFINE_EVENT(smc_msg_event, smc_tx_sendmsg,
+
+	     TP_PROTO(const struct smc_sock *smc, size_t len),
+
+	     TP_ARGS(smc, len)
+);
+
+DEFINE_EVENT(smc_msg_event, smc_rx_recvmsg,
+
+	     TP_PROTO(const struct smc_sock *smc, size_t len),
+
+	     TP_ARGS(smc, len)
+);
+
+TRACE_EVENT(smcr_link_down,
+
+	    TP_PROTO(const struct smc_link *lnk, void *location),
+
+	    TP_ARGS(lnk, location),
+
+	    TP_STRUCT__entry(
+			     __field(const void *, lnk)
+			     __field(const void *, lgr)
+			     __field(u64, net_cookie)
+			     __field(int, state)
+			     __string(name, lnk->ibname)
+			     __field(void *, location)
+	    ),
+
+	    TP_fast_assign(
+			   const struct smc_link_group *lgr = lnk->lgr;
+
+			   __entry->lnk = lnk;
+			   __entry->lgr = lgr;
+			   __entry->net_cookie = lgr->net->net_cookie;
+			   __entry->state = lnk->state;
+			   __assign_str(name);
+			   __entry->location = location;
+	    ),
+
+	    TP_printk("lnk=%p lgr=%p net=%llu state=%d dev=%s location=%pS",
+		      __entry->lnk, __entry->lgr, __entry->net_cookie,
+		      __entry->state, __get_str(name),
+		      __entry->location)
+);
+
+#endif /* _TRACE_SMC_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE smc_tracepoint
+
+#include <trace/define_trace.h>

linux-7.2.3/net/smc/smc_tx.c

@@ -0,0 +1,737 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Manage send buffer.
+ * Producer:
+ * Copy user space data into send buffer, if send buffer space available.
+ * Consumer:
+ * Trigger RDMA write into RMBE of peer and send CDC, if RMBE space available.
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#include <linux/net.h>
+#include <linux/rcupdate.h>
+#include <linux/workqueue.h>
+#include <linux/sched/signal.h>
+
+#include <net/sock.h>
+#include <net/tcp.h>
+
+#include "smc.h"
+#include "smc_wr.h"
+#include "smc_cdc.h"
+#include "smc_close.h"
+#include "smc_ism.h"
+#include "smc_tx.h"
+#include "smc_stats.h"
+#include "smc_tracepoint.h"
+
+#define SMC_TX_WORK_DELAY	0
+
+/***************************** sndbuf producer *******************************/
+
+/* callback implementation for sk.sk_write_space()
+ * to wakeup sndbuf producers that blocked with smc_tx_wait().
+ * called under sk_socket lock.
+ */
+static void smc_tx_write_space(struct sock *sk)
+{
+	struct socket *sock = sk->sk_socket;
+	struct smc_sock *smc = smc_sk(sk);
+	struct socket_wq *wq;
+
+	/* 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, !smc->conn.lnk);
+		clear_bit(SOCK_NOSPACE, &sock->flags);
+		rcu_read_lock();
+		wq = rcu_dereference(sk->sk_wq);
+		if (skwq_has_sleeper(wq))
+			wake_up_interruptible_poll(&wq->wait,
+						   EPOLLOUT | EPOLLWRNORM |
+						   EPOLLWRBAND);
+		if (wq && wq->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN))
+			sock_wake_async(wq, SOCK_WAKE_SPACE, POLL_OUT);
+		rcu_read_unlock();
+	}
+}
+
+/* Wakeup sndbuf producers that blocked with smc_tx_wait().
+ * Cf. tcp_data_snd_check()=>tcp_check_space()=>tcp_new_space().
+ */
+void smc_tx_sndbuf_nonfull(struct smc_sock *smc)
+{
+	if (smc->sk.sk_socket &&
+	    test_bit(SOCK_NOSPACE, &smc->sk.sk_socket->flags))
+		smc->sk.sk_write_space(&smc->sk);
+}
+
+/* blocks sndbuf producer until at least one byte of free space available
+ * or urgent Byte was consumed
+ */
+static int smc_tx_wait(struct smc_sock *smc, int flags)
+{
+	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	struct smc_connection *conn = &smc->conn;
+	struct sock *sk = &smc->sk;
+	long timeo;
+	int rc = 0;
+
+	/* similar to sk_stream_wait_memory */
+	timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
+	add_wait_queue(sk_sleep(sk), &wait);
+	while (1) {
+		sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+		if (sk->sk_err ||
+		    (sk->sk_shutdown & SEND_SHUTDOWN) ||
+		    conn->killed ||
+		    conn->local_tx_ctrl.conn_state_flags.peer_done_writing) {
+			rc = -EPIPE;
+			break;
+		}
+		if (smc_cdc_rxed_any_close(conn)) {
+			rc = -ECONNRESET;
+			break;
+		}
+		if (!timeo) {
+			/* ensure EPOLLOUT is subsequently generated */
+			set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+			rc = -EAGAIN;
+			break;
+		}
+		if (signal_pending(current)) {
+			rc = sock_intr_errno(timeo);
+			break;
+		}
+		sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+		if (atomic_read(&conn->sndbuf_space) && !conn->urg_tx_pend)
+			break; /* at least 1 byte of free & no urgent data */
+		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+		sk_wait_event(sk, &timeo,
+			      READ_ONCE(sk->sk_err) ||
+			      (READ_ONCE(sk->sk_shutdown) & SEND_SHUTDOWN) ||
+			      smc_cdc_rxed_any_close(conn) ||
+			      (atomic_read(&conn->sndbuf_space) &&
+			       !conn->urg_tx_pend),
+			      &wait);
+	}
+	remove_wait_queue(sk_sleep(sk), &wait);
+	return rc;
+}
+
+static bool smc_tx_is_corked(struct smc_sock *smc)
+{
+	struct tcp_sock *tp = tcp_sk(smc->clcsock->sk);
+
+	return (tp->nonagle & TCP_NAGLE_CORK) ? true : false;
+}
+
+/* If we have pending CDC messages, do not send:
+ * Because CQE of this CDC message will happen shortly, it gives
+ * a chance to coalesce future sendmsg() payload in to one RDMA Write,
+ * without need for a timer, and with no latency trade off.
+ * Algorithm here:
+ *  1. First message should never cork
+ *  2. If we have pending Tx CDC messages, wait for the first CDC
+ *     message's completion
+ *  3. Don't cork to much data in a single RDMA Write to prevent burst
+ *     traffic, total corked message should not exceed sendbuf/2
+ */
+static bool smc_should_autocork(struct smc_sock *smc)
+{
+	struct smc_connection *conn = &smc->conn;
+	int corking_size;
+
+	corking_size = min_t(unsigned int, conn->sndbuf_desc->len >> 1,
+			     sock_net(&smc->sk)->smc.sysctl_autocorking_size);
+
+	if (atomic_read(&conn->cdc_pend_tx_wr) == 0 ||
+	    smc_tx_prepared_sends(conn) > corking_size)
+		return false;
+	return true;
+}
+
+static bool smc_tx_should_cork(struct smc_sock *smc, struct msghdr *msg)
+{
+	struct smc_connection *conn = &smc->conn;
+
+	if (smc_should_autocork(smc))
+		return true;
+
+	/* for a corked socket defer the RDMA writes if
+	 * sndbuf_space is still available. The applications
+	 * should known how/when to uncork it.
+	 */
+	if ((msg->msg_flags & MSG_MORE ||
+	     smc_tx_is_corked(smc)) &&
+	    atomic_read(&conn->sndbuf_space))
+		return true;
+
+	return false;
+}
+
+/* sndbuf producer: main API called by socket layer.
+ * called under sock lock.
+ */
+int smc_tx_sendmsg(struct smc_sock *smc, struct msghdr *msg, size_t len)
+{
+	size_t copylen, send_done = 0, send_remaining = len;
+	size_t chunk_len, chunk_off, chunk_len_sum;
+	struct smc_connection *conn = &smc->conn;
+	union smc_host_cursor prep;
+	struct sock *sk = &smc->sk;
+	char *sndbuf_base;
+	int tx_cnt_prep;
+	int writespace;
+	int rc, chunk;
+
+	/* This should be in poll */
+	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+
+	if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) {
+		rc = -EPIPE;
+		goto out_err;
+	}
+
+	if (sk->sk_state == SMC_INIT)
+		return -ENOTCONN;
+
+	if (len > conn->sndbuf_desc->len)
+		SMC_STAT_RMB_TX_SIZE_SMALL(smc, !conn->lnk);
+
+	if (len > conn->peer_rmbe_size)
+		SMC_STAT_RMB_TX_PEER_SIZE_SMALL(smc, !conn->lnk);
+
+	if (msg->msg_flags & MSG_OOB)
+		SMC_STAT_INC(smc, urg_data_cnt);
+
+	while (msg_data_left(msg)) {
+		if (smc->sk.sk_shutdown & SEND_SHUTDOWN ||
+		    (smc->sk.sk_err == ECONNABORTED) ||
+		    conn->killed)
+			return -EPIPE;
+		if (smc_cdc_rxed_any_close(conn))
+			return send_done ?: -ECONNRESET;
+
+		if (msg->msg_flags & MSG_OOB)
+			conn->local_tx_ctrl.prod_flags.urg_data_pending = 1;
+
+		if (!atomic_read(&conn->sndbuf_space) || conn->urg_tx_pend) {
+			if (send_done)
+				return send_done;
+			rc = smc_tx_wait(smc, msg->msg_flags);
+			if (rc)
+				goto out_err;
+			continue;
+		}
+
+		/* initialize variables for 1st iteration of subsequent loop */
+		/* could be just 1 byte, even after smc_tx_wait above */
+		writespace = atomic_read(&conn->sndbuf_space);
+		/* not more than what user space asked for */
+		copylen = min_t(size_t, send_remaining, writespace);
+		/* determine start of sndbuf */
+		sndbuf_base = conn->sndbuf_desc->cpu_addr;
+		smc_curs_copy(&prep, &conn->tx_curs_prep, conn);
+		tx_cnt_prep = prep.count;
+		/* determine chunks where to write into sndbuf */
+		/* either unwrapped case, or 1st chunk of wrapped case */
+		chunk_len = min_t(size_t, copylen, conn->sndbuf_desc->len -
+				  tx_cnt_prep);
+		chunk_len_sum = chunk_len;
+		chunk_off = tx_cnt_prep;
+		for (chunk = 0; chunk < 2; chunk++) {
+			rc = memcpy_from_msg(sndbuf_base + chunk_off,
+					     msg, chunk_len);
+			if (rc) {
+				smc_sndbuf_sync_sg_for_device(conn);
+				if (send_done)
+					return send_done;
+				goto out_err;
+			}
+			send_done += chunk_len;
+			send_remaining -= chunk_len;
+
+			if (chunk_len_sum == copylen)
+				break; /* either on 1st or 2nd iteration */
+			/* prepare next (== 2nd) iteration */
+			chunk_len = copylen - chunk_len; /* remainder */
+			chunk_len_sum += chunk_len;
+			chunk_off = 0; /* modulo offset in send ring buffer */
+		}
+		smc_sndbuf_sync_sg_for_device(conn);
+		/* update cursors */
+		smc_curs_add(conn->sndbuf_desc->len, &prep, copylen);
+		smc_curs_copy(&conn->tx_curs_prep, &prep, conn);
+		/* increased in send tasklet smc_cdc_tx_handler() */
+		smp_mb__before_atomic();
+		atomic_sub(copylen, &conn->sndbuf_space);
+		/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
+		smp_mb__after_atomic();
+		/* since we just produced more new data into sndbuf,
+		 * trigger sndbuf consumer: RDMA write into peer RMBE and CDC
+		 */
+		if ((msg->msg_flags & MSG_OOB) && !send_remaining)
+			conn->urg_tx_pend = true;
+		/* If we need to cork, do nothing and wait for the next
+		 * sendmsg() call or push on tx completion
+		 */
+		if (!smc_tx_should_cork(smc, msg))
+			smc_tx_sndbuf_nonempty(conn);
+
+		trace_smc_tx_sendmsg(smc, copylen);
+	} /* while (msg_data_left(msg)) */
+
+	return send_done;
+
+out_err:
+	rc = sk_stream_error(sk, msg->msg_flags, rc);
+	/* make sure we wake any epoll edge trigger waiter */
+	if (unlikely(rc == -EAGAIN))
+		sk->sk_write_space(sk);
+	return rc;
+}
+
+/***************************** sndbuf consumer *******************************/
+
+/* sndbuf consumer: actual data transfer of one target chunk with ISM write */
+int smcd_tx_ism_write(struct smc_connection *conn, void *data, size_t len,
+		      u32 offset, int signal)
+{
+	int rc;
+
+	rc = smc_ism_write(conn->lgr->smcd, conn->peer_token,
+			   conn->peer_rmbe_idx, signal, conn->tx_off + offset,
+			   data, len);
+	if (rc)
+		conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
+	return rc;
+}
+
+/* sndbuf consumer: actual data transfer of one target chunk with RDMA write */
+static int smc_tx_rdma_write(struct smc_connection *conn, int peer_rmbe_offset,
+			     int num_sges, struct ib_rdma_wr *rdma_wr)
+{
+	struct smc_link_group *lgr = conn->lgr;
+	struct smc_link *link = conn->lnk;
+	int rc;
+
+	rdma_wr->wr.wr_id = smc_wr_tx_get_next_wr_id(link);
+	rdma_wr->wr.num_sge = num_sges;
+	rdma_wr->remote_addr =
+		lgr->rtokens[conn->rtoken_idx][link->link_idx].dma_addr +
+		/* RMBE within RMB */
+		conn->tx_off +
+		/* offset within RMBE */
+		peer_rmbe_offset;
+	rdma_wr->rkey = lgr->rtokens[conn->rtoken_idx][link->link_idx].rkey;
+	rc = ib_post_send(link->roce_qp, &rdma_wr->wr, NULL);
+	if (rc)
+		smcr_link_down_cond_sched(link);
+	return rc;
+}
+
+/* sndbuf consumer */
+static inline void smc_tx_advance_cursors(struct smc_connection *conn,
+					  union smc_host_cursor *prod,
+					  union smc_host_cursor *sent,
+					  size_t len)
+{
+	smc_curs_add(conn->peer_rmbe_size, prod, len);
+	/* increased in recv tasklet smc_cdc_msg_rcv() */
+	smp_mb__before_atomic();
+	/* data in flight reduces usable snd_wnd */
+	atomic_sub(len, &conn->peer_rmbe_space);
+	/* guarantee 0 <= peer_rmbe_space <= peer_rmbe_size */
+	smp_mb__after_atomic();
+	smc_curs_add(conn->sndbuf_desc->len, sent, len);
+}
+
+/* SMC-R helper for smc_tx_rdma_writes() */
+static int smcr_tx_rdma_writes(struct smc_connection *conn, size_t len,
+			       size_t src_off, size_t src_len,
+			       size_t dst_off, size_t dst_len,
+			       struct smc_rdma_wr *wr_rdma_buf)
+{
+	struct smc_link *link = conn->lnk;
+
+	dma_addr_t dma_addr =
+		sg_dma_address(conn->sndbuf_desc->sgt[link->link_idx].sgl);
+	u64 virt_addr = (uintptr_t)conn->sndbuf_desc->cpu_addr;
+	int src_len_sum = src_len, dst_len_sum = dst_len;
+	int sent_count = src_off;
+	int srcchunk, dstchunk;
+	int num_sges;
+	int rc;
+
+	for (dstchunk = 0; dstchunk < 2; dstchunk++) {
+		struct ib_rdma_wr *wr = &wr_rdma_buf->wr_tx_rdma[dstchunk];
+		struct ib_sge *sge = wr->wr.sg_list;
+		u64 base_addr = dma_addr;
+
+		if (dst_len < link->qp_attr.cap.max_inline_data) {
+			base_addr = virt_addr;
+			wr->wr.send_flags |= IB_SEND_INLINE;
+		} else {
+			wr->wr.send_flags &= ~IB_SEND_INLINE;
+		}
+
+		num_sges = 0;
+		for (srcchunk = 0; srcchunk < 2; srcchunk++) {
+			sge[srcchunk].addr = conn->sndbuf_desc->is_vm ?
+				(virt_addr + src_off) : (base_addr + src_off);
+			sge[srcchunk].length = src_len;
+			if (conn->sndbuf_desc->is_vm)
+				sge[srcchunk].lkey =
+					conn->sndbuf_desc->mr[link->link_idx]->lkey;
+			num_sges++;
+
+			src_off += src_len;
+			if (src_off >= conn->sndbuf_desc->len)
+				src_off -= conn->sndbuf_desc->len;
+						/* modulo in send ring */
+			if (src_len_sum == dst_len)
+				break; /* either on 1st or 2nd iteration */
+			/* prepare next (== 2nd) iteration */
+			src_len = dst_len - src_len; /* remainder */
+			src_len_sum += src_len;
+		}
+		rc = smc_tx_rdma_write(conn, dst_off, num_sges, wr);
+		if (rc)
+			return rc;
+		if (dst_len_sum == len)
+			break; /* either on 1st or 2nd iteration */
+		/* prepare next (== 2nd) iteration */
+		dst_off = 0; /* modulo offset in RMBE ring buffer */
+		dst_len = len - dst_len; /* remainder */
+		dst_len_sum += dst_len;
+		src_len = min_t(int, dst_len, conn->sndbuf_desc->len -
+				sent_count);
+		src_len_sum = src_len;
+	}
+	return 0;
+}
+
+/* SMC-D helper for smc_tx_rdma_writes() */
+static int smcd_tx_rdma_writes(struct smc_connection *conn, size_t len,
+			       size_t src_off, size_t src_len,
+			       size_t dst_off, size_t dst_len)
+{
+	int src_len_sum = src_len, dst_len_sum = dst_len;
+	int srcchunk, dstchunk;
+	int rc;
+
+	if (conn->sndbuf_desc->is_attached)
+		return 0;
+
+	for (dstchunk = 0; dstchunk < 2; dstchunk++) {
+		for (srcchunk = 0; srcchunk < 2; srcchunk++) {
+			void *data = conn->sndbuf_desc->cpu_addr + src_off;
+
+			rc = smcd_tx_ism_write(conn, data, src_len, dst_off +
+					       sizeof(struct smcd_cdc_msg), 0);
+			if (rc)
+				return rc;
+			dst_off += src_len;
+			src_off += src_len;
+			if (src_off >= conn->sndbuf_desc->len)
+				src_off -= conn->sndbuf_desc->len;
+						/* modulo in send ring */
+			if (src_len_sum == dst_len)
+				break; /* either on 1st or 2nd iteration */
+			/* prepare next (== 2nd) iteration */
+			src_len = dst_len - src_len; /* remainder */
+			src_len_sum += src_len;
+		}
+		if (dst_len_sum == len)
+			break; /* either on 1st or 2nd iteration */
+		/* prepare next (== 2nd) iteration */
+		dst_off = 0; /* modulo offset in RMBE ring buffer */
+		dst_len = len - dst_len; /* remainder */
+		dst_len_sum += dst_len;
+		src_len = min_t(int, dst_len, conn->sndbuf_desc->len - src_off);
+		src_len_sum = src_len;
+	}
+	return 0;
+}
+
+/* sndbuf consumer: prepare all necessary (src&dst) chunks of data transmit;
+ * usable snd_wnd as max transmit
+ */
+static int smc_tx_rdma_writes(struct smc_connection *conn,
+			      struct smc_rdma_wr *wr_rdma_buf)
+{
+	size_t len, src_len, dst_off, dst_len; /* current chunk values */
+	union smc_host_cursor sent, prep, prod, cons;
+	struct smc_cdc_producer_flags *pflags;
+	int to_send, rmbespace;
+	int rc;
+
+	/* source: sndbuf */
+	smc_curs_copy(&sent, &conn->tx_curs_sent, conn);
+	smc_curs_copy(&prep, &conn->tx_curs_prep, conn);
+	/* cf. wmem_alloc - (snd_max - snd_una) */
+	to_send = smc_curs_diff(conn->sndbuf_desc->len, &sent, &prep);
+	if (to_send <= 0)
+		return 0;
+
+	/* destination: RMBE */
+	/* cf. snd_wnd */
+	rmbespace = atomic_read(&conn->peer_rmbe_space);
+	if (rmbespace <= 0) {
+		struct smc_sock *smc = container_of(conn, struct smc_sock,
+						    conn);
+		SMC_STAT_RMB_TX_PEER_FULL(smc, !conn->lnk);
+		return 0;
+	}
+	smc_curs_copy(&prod, &conn->local_tx_ctrl.prod, conn);
+	smc_curs_copy(&cons, &conn->local_rx_ctrl.cons, conn);
+
+	/* if usable snd_wnd closes ask peer to advertise once it opens again */
+	pflags = &conn->local_tx_ctrl.prod_flags;
+	pflags->write_blocked = (to_send >= rmbespace);
+	/* cf. usable snd_wnd */
+	len = min(to_send, rmbespace);
+
+	/* initialize variables for first iteration of subsequent nested loop */
+	dst_off = prod.count;
+	if (prod.wrap == cons.wrap) {
+		/* the filled destination area is unwrapped,
+		 * hence the available free destination space is wrapped
+		 * and we need 2 destination chunks of sum len; start with 1st
+		 * which is limited by what's available in sndbuf
+		 */
+		dst_len = min_t(size_t,
+				conn->peer_rmbe_size - prod.count, len);
+	} else {
+		/* the filled destination area is wrapped,
+		 * hence the available free destination space is unwrapped
+		 * and we need a single destination chunk of entire len
+		 */
+		dst_len = len;
+	}
+	/* dst_len determines the maximum src_len */
+	if (sent.count + dst_len <= conn->sndbuf_desc->len) {
+		/* unwrapped src case: single chunk of entire dst_len */
+		src_len = dst_len;
+	} else {
+		/* wrapped src case: 2 chunks of sum dst_len; start with 1st: */
+		src_len = conn->sndbuf_desc->len - sent.count;
+	}
+
+	if (conn->lgr->is_smcd)
+		rc = smcd_tx_rdma_writes(conn, len, sent.count, src_len,
+					 dst_off, dst_len);
+	else
+		rc = smcr_tx_rdma_writes(conn, len, sent.count, src_len,
+					 dst_off, dst_len, wr_rdma_buf);
+	if (rc)
+		return rc;
+
+	if (conn->urg_tx_pend && len == to_send)
+		pflags->urg_data_present = 1;
+	smc_tx_advance_cursors(conn, &prod, &sent, len);
+	/* update connection's cursors with advanced local cursors */
+	smc_curs_copy(&conn->local_tx_ctrl.prod, &prod, conn);
+							/* dst: peer RMBE */
+	smc_curs_copy(&conn->tx_curs_sent, &sent, conn);/* src: local sndbuf */
+
+	return 0;
+}
+
+/* Wakeup sndbuf consumers from any context (IRQ or process)
+ * since there is more data to transmit; usable snd_wnd as max transmit
+ */
+static int smcr_tx_sndbuf_nonempty(struct smc_connection *conn)
+{
+	struct smc_cdc_producer_flags *pflags = &conn->local_tx_ctrl.prod_flags;
+	struct smc_link *link = conn->lnk;
+	struct smc_rdma_wr *wr_rdma_buf;
+	struct smc_cdc_tx_pend *pend;
+	struct smc_wr_buf *wr_buf;
+	int rc;
+
+	if (!link || !smc_wr_tx_link_hold(link))
+		return -ENOLINK;
+	rc = smc_cdc_get_free_slot(conn, link, &wr_buf, &wr_rdma_buf, &pend);
+	if (rc < 0) {
+		smc_wr_tx_link_put(link);
+		if (rc == -EBUSY) {
+			struct smc_sock *smc =
+				container_of(conn, struct smc_sock, conn);
+
+			if (smc->sk.sk_err == ECONNABORTED)
+				return sock_error(&smc->sk);
+			if (conn->killed)
+				return -EPIPE;
+			rc = 0;
+			mod_delayed_work(conn->lgr->tx_wq, &conn->tx_work,
+					 SMC_TX_WORK_DELAY);
+		}
+		return rc;
+	}
+
+	spin_lock_bh(&conn->send_lock);
+	if (link != conn->lnk) {
+		/* link of connection changed, tx_work will restart */
+		smc_wr_tx_put_slot(link,
+				   (struct smc_wr_tx_pend_priv *)pend);
+		rc = -ENOLINK;
+		goto out_unlock;
+	}
+	if (!pflags->urg_data_present) {
+		rc = smc_tx_rdma_writes(conn, wr_rdma_buf);
+		if (rc) {
+			smc_wr_tx_put_slot(link,
+					   (struct smc_wr_tx_pend_priv *)pend);
+			goto out_unlock;
+		}
+	}
+
+	rc = smc_cdc_msg_send(conn, wr_buf, pend);
+	if (!rc && pflags->urg_data_present) {
+		pflags->urg_data_pending = 0;
+		pflags->urg_data_present = 0;
+	}
+
+out_unlock:
+	spin_unlock_bh(&conn->send_lock);
+	smc_wr_tx_link_put(link);
+	return rc;
+}
+
+static int smcd_tx_sndbuf_nonempty(struct smc_connection *conn)
+{
+	struct smc_cdc_producer_flags *pflags = &conn->local_tx_ctrl.prod_flags;
+	int rc = 0;
+
+	spin_lock_bh(&conn->send_lock);
+	if (!pflags->urg_data_present)
+		rc = smc_tx_rdma_writes(conn, NULL);
+	if (!rc)
+		rc = smcd_cdc_msg_send(conn);
+
+	if (!rc && pflags->urg_data_present) {
+		pflags->urg_data_pending = 0;
+		pflags->urg_data_present = 0;
+	}
+	spin_unlock_bh(&conn->send_lock);
+	return rc;
+}
+
+int smc_tx_sndbuf_nonempty(struct smc_connection *conn)
+{
+	struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
+	int rc = 0;
+
+	/* No data in the send queue */
+	if (unlikely(smc_tx_prepared_sends(conn) <= 0))
+		goto out;
+
+	/* Peer don't have RMBE space */
+	if (unlikely(atomic_read(&conn->peer_rmbe_space) <= 0)) {
+		SMC_STAT_RMB_TX_PEER_FULL(smc, !conn->lnk);
+		goto out;
+	}
+
+	if (conn->killed ||
+	    conn->local_rx_ctrl.conn_state_flags.peer_conn_abort) {
+		rc = -EPIPE;    /* connection being aborted */
+		goto out;
+	}
+	if (conn->lgr->is_smcd)
+		rc = smcd_tx_sndbuf_nonempty(conn);
+	else
+		rc = smcr_tx_sndbuf_nonempty(conn);
+
+	if (!rc) {
+		/* trigger socket release if connection is closing */
+		smc_close_wake_tx_prepared(smc);
+	}
+
+out:
+	return rc;
+}
+
+/* Wakeup sndbuf consumers from process context
+ * since there is more data to transmit. The caller
+ * must hold sock lock.
+ */
+void smc_tx_pending(struct smc_connection *conn)
+{
+	struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
+	int rc;
+
+	if (smc->sk.sk_err)
+		return;
+
+	rc = smc_tx_sndbuf_nonempty(conn);
+	if (!rc && conn->local_rx_ctrl.prod_flags.write_blocked &&
+	    !atomic_read(&conn->bytes_to_rcv))
+		conn->local_rx_ctrl.prod_flags.write_blocked = 0;
+}
+
+/* Wakeup sndbuf consumers from process context
+ * since there is more data to transmit in locked
+ * sock.
+ */
+void smc_tx_work(struct work_struct *work)
+{
+	struct smc_connection *conn = container_of(to_delayed_work(work),
+						   struct smc_connection,
+						   tx_work);
+	struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
+
+	lock_sock(&smc->sk);
+	smc_tx_pending(conn);
+	release_sock(&smc->sk);
+}
+
+void smc_tx_consumer_update(struct smc_connection *conn, bool force)
+{
+	union smc_host_cursor cfed, cons, prod;
+	int sender_free = conn->rmb_desc->len;
+	int to_confirm;
+
+	smc_curs_copy(&cons, &conn->local_tx_ctrl.cons, conn);
+	smc_curs_copy(&cfed, &conn->rx_curs_confirmed, conn);
+	to_confirm = smc_curs_diff(conn->rmb_desc->len, &cfed, &cons);
+	if (to_confirm > conn->rmbe_update_limit) {
+		smc_curs_copy(&prod, &conn->local_rx_ctrl.prod, conn);
+		sender_free = conn->rmb_desc->len -
+			      smc_curs_diff_large(conn->rmb_desc->len,
+						  &cfed, &prod);
+	}
+
+	if (conn->local_rx_ctrl.prod_flags.cons_curs_upd_req ||
+	    force ||
+	    ((to_confirm > conn->rmbe_update_limit) &&
+	     ((sender_free <= (conn->rmb_desc->len / 2)) ||
+	      conn->local_rx_ctrl.prod_flags.write_blocked))) {
+		if (conn->killed ||
+		    conn->local_rx_ctrl.conn_state_flags.peer_conn_abort)
+			return;
+		if ((smc_cdc_get_slot_and_msg_send(conn) < 0) &&
+		    !conn->killed) {
+			queue_delayed_work(conn->lgr->tx_wq, &conn->tx_work,
+					   SMC_TX_WORK_DELAY);
+			return;
+		}
+	}
+	if (conn->local_rx_ctrl.prod_flags.write_blocked &&
+	    !atomic_read(&conn->bytes_to_rcv))
+		conn->local_rx_ctrl.prod_flags.write_blocked = 0;
+}
+
+/***************************** send initialize *******************************/
+
+/* Initialize send properties on connection establishment. NB: not __init! */
+void smc_tx_init(struct smc_sock *smc)
+{
+	smc->sk.sk_write_space = smc_tx_write_space;
+}

linux-7.2.3/net/smc/smc_tx.h

@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Manage send buffer
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Ursula Braun <ubraun@linux.vnet.ibm.com>
+ */
+
+#ifndef SMC_TX_H
+#define SMC_TX_H
+
+#include <linux/socket.h>
+#include <linux/types.h>
+
+#include "smc.h"
+#include "smc_cdc.h"
+
+static inline int smc_tx_prepared_sends(struct smc_connection *conn)
+{
+	union smc_host_cursor sent, prep;
+
+	smc_curs_copy(&sent, &conn->tx_curs_sent, conn);
+	smc_curs_copy(&prep, &conn->tx_curs_prep, conn);
+	return smc_curs_diff(conn->sndbuf_desc->len, &sent, &prep);
+}
+
+void smc_tx_pending(struct smc_connection *conn);
+void smc_tx_work(struct work_struct *work);
+void smc_tx_init(struct smc_sock *smc);
+int smc_tx_sendmsg(struct smc_sock *smc, struct msghdr *msg, size_t len);
+int smc_tx_sndbuf_nonempty(struct smc_connection *conn);
+void smc_tx_sndbuf_nonfull(struct smc_sock *smc);
+void smc_tx_consumer_update(struct smc_connection *conn, bool force);
+int smcd_tx_ism_write(struct smc_connection *conn, void *data, size_t len,
+		      u32 offset, int signal);
+
+#endif /* SMC_TX_H */

linux-7.2.3/net/smc/smc_wr.c

@@ -0,0 +1,933 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Work Requests exploiting Infiniband API
+ *
+ * Work requests (WR) of type ib_post_send or ib_post_recv respectively
+ * are submitted to either RC SQ or RC RQ respectively
+ * (reliably connected send/receive queue)
+ * and become work queue entries (WQEs).
+ * While an SQ WR/WQE is pending, we track it until transmission completion.
+ * Through a send or receive completion queue (CQ) respectively,
+ * we get completion queue entries (CQEs) [aka work completions (WCs)].
+ * Since the CQ callback is called from IRQ context, we split work by using
+ * bottom halves implemented by tasklets.
+ *
+ * SMC uses this to exchange LLC (link layer control)
+ * and CDC (connection data control) messages.
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Steffen Maier <maier@linux.vnet.ibm.com>
+ */
+
+#include <linux/atomic.h>
+#include <linux/hashtable.h>
+#include <linux/wait.h>
+#include <rdma/ib_verbs.h>
+#include <asm/div64.h>
+
+#include "smc.h"
+#include "smc_wr.h"
+
+#define SMC_WR_MAX_POLL_CQE 10	/* max. # of compl. queue elements in 1 poll */
+
+#define SMC_WR_RX_HASH_BITS 4
+static DEFINE_HASHTABLE(smc_wr_rx_hash, SMC_WR_RX_HASH_BITS);
+static DEFINE_SPINLOCK(smc_wr_rx_hash_lock);
+
+struct smc_wr_tx_pend {	/* control data for a pending send request */
+	u64			wr_id;		/* work request id sent */
+	smc_wr_tx_handler	handler;
+	enum ib_wc_status	wc_status;	/* CQE status */
+	struct smc_link		*link;
+	u32			idx;
+	struct smc_wr_tx_pend_priv priv;
+	u8			compl_requested;
+};
+
+/******************************** send queue *********************************/
+
+/*------------------------------- completion --------------------------------*/
+
+/* returns true if at least one tx work request is pending on the given link */
+static inline bool smc_wr_is_tx_pend(struct smc_link *link)
+{
+	return !bitmap_empty(link->wr_tx_mask, link->wr_tx_cnt);
+}
+
+/* wait till all pending tx work requests on the given link are completed */
+void smc_wr_tx_wait_no_pending_sends(struct smc_link *link)
+{
+	wait_event(link->wr_tx_wait, !smc_wr_is_tx_pend(link));
+}
+
+static inline int smc_wr_tx_find_pending_index(struct smc_link *link, u64 wr_id)
+{
+	u32 i;
+
+	for (i = 0; i < link->wr_tx_cnt; i++) {
+		if (link->wr_tx_pends[i].wr_id == wr_id)
+			return i;
+	}
+	return link->wr_tx_cnt;
+}
+
+static inline void smc_wr_tx_process_cqe(struct ib_wc *wc)
+{
+	struct smc_wr_tx_pend pnd_snd;
+	struct smc_link *link;
+	u32 pnd_snd_idx;
+
+	link = wc->qp->qp_context;
+
+	if (wc->opcode == IB_WC_REG_MR) {
+		if (wc->status)
+			link->wr_reg_state = FAILED;
+		else
+			link->wr_reg_state = CONFIRMED;
+		smc_wr_wakeup_reg_wait(link);
+		return;
+	}
+
+	pnd_snd_idx = smc_wr_tx_find_pending_index(link, wc->wr_id);
+	if (pnd_snd_idx == link->wr_tx_cnt) {
+		if (link->lgr->smc_version != SMC_V2 ||
+		    link->wr_tx_v2_pend->wr_id != wc->wr_id)
+			return;
+		link->wr_tx_v2_pend->wc_status = wc->status;
+		memcpy(&pnd_snd, link->wr_tx_v2_pend, sizeof(pnd_snd));
+		/* clear the full struct smc_wr_tx_pend including .priv */
+		memset(link->wr_tx_v2_pend, 0,
+		       sizeof(*link->wr_tx_v2_pend));
+		memset(link->lgr->wr_tx_buf_v2, 0,
+		       sizeof(*link->lgr->wr_tx_buf_v2));
+	} else {
+		link->wr_tx_pends[pnd_snd_idx].wc_status = wc->status;
+		if (link->wr_tx_pends[pnd_snd_idx].compl_requested)
+			complete(&link->wr_tx_compl[pnd_snd_idx]);
+		memcpy(&pnd_snd, &link->wr_tx_pends[pnd_snd_idx],
+		       sizeof(pnd_snd));
+		/* clear the full struct smc_wr_tx_pend including .priv */
+		memset(&link->wr_tx_pends[pnd_snd_idx], 0,
+		       sizeof(link->wr_tx_pends[pnd_snd_idx]));
+		memset(&link->wr_tx_bufs[pnd_snd_idx], 0,
+		       sizeof(link->wr_tx_bufs[pnd_snd_idx]));
+		if (!test_and_clear_bit(pnd_snd_idx, link->wr_tx_mask))
+			return;
+	}
+
+	if (wc->status) {
+		if (link->lgr->smc_version == SMC_V2) {
+			memset(link->wr_tx_v2_pend, 0,
+			       sizeof(*link->wr_tx_v2_pend));
+			memset(link->lgr->wr_tx_buf_v2, 0,
+			       sizeof(*link->lgr->wr_tx_buf_v2));
+		}
+		/* terminate link */
+		smcr_link_down_cond_sched(link);
+	}
+	if (pnd_snd.handler)
+		pnd_snd.handler(&pnd_snd.priv, link, wc->status);
+	wake_up(&link->wr_tx_wait);
+}
+
+static void smc_wr_tx_tasklet_fn(struct tasklet_struct *t)
+{
+	struct smc_ib_device *dev = from_tasklet(dev, t, send_tasklet);
+	struct ib_wc wc[SMC_WR_MAX_POLL_CQE];
+	int i = 0, rc;
+	int polled = 0;
+
+again:
+	polled++;
+	do {
+		memset(&wc, 0, sizeof(wc));
+		rc = ib_poll_cq(dev->roce_cq_send, SMC_WR_MAX_POLL_CQE, wc);
+		if (polled == 1) {
+			ib_req_notify_cq(dev->roce_cq_send,
+					 IB_CQ_NEXT_COMP |
+					 IB_CQ_REPORT_MISSED_EVENTS);
+		}
+		if (!rc)
+			break;
+		for (i = 0; i < rc; i++)
+			smc_wr_tx_process_cqe(&wc[i]);
+	} while (rc > 0);
+	if (polled == 1)
+		goto again;
+}
+
+void smc_wr_tx_cq_handler(struct ib_cq *ib_cq, void *cq_context)
+{
+	struct smc_ib_device *dev = (struct smc_ib_device *)cq_context;
+
+	tasklet_schedule(&dev->send_tasklet);
+}
+
+/*---------------------------- request submission ---------------------------*/
+
+static inline int smc_wr_tx_get_free_slot_index(struct smc_link *link, u32 *idx)
+{
+	*idx = link->wr_tx_cnt;
+	if (!smc_link_sendable(link))
+		return -ENOLINK;
+	for_each_clear_bit(*idx, link->wr_tx_mask, link->wr_tx_cnt) {
+		if (!test_and_set_bit(*idx, link->wr_tx_mask))
+			return 0;
+	}
+	*idx = link->wr_tx_cnt;
+	return -EBUSY;
+}
+
+/**
+ * smc_wr_tx_get_free_slot() - returns buffer for message assembly,
+ *			and sets info for pending transmit tracking
+ * @link:		Pointer to smc_link used to later send the message.
+ * @handler:		Send completion handler function pointer.
+ * @wr_buf:		Out value returns pointer to message buffer.
+ * @wr_rdma_buf:	Out value returns pointer to rdma work request.
+ * @wr_pend_priv:	Out value returns pointer serving as handler context.
+ *
+ * Return: 0 on success, or -errno on error.
+ */
+int smc_wr_tx_get_free_slot(struct smc_link *link,
+			    smc_wr_tx_handler handler,
+			    struct smc_wr_buf **wr_buf,
+			    struct smc_rdma_wr **wr_rdma_buf,
+			    struct smc_wr_tx_pend_priv **wr_pend_priv)
+{
+	struct smc_link_group *lgr = smc_get_lgr(link);
+	struct smc_wr_tx_pend *wr_pend;
+	u32 idx = link->wr_tx_cnt;
+	struct ib_send_wr *wr_ib;
+	u64 wr_id;
+	int rc;
+
+	*wr_buf = NULL;
+	*wr_pend_priv = NULL;
+	if (in_softirq() || lgr->terminating) {
+		rc = smc_wr_tx_get_free_slot_index(link, &idx);
+		if (rc)
+			return rc;
+	} else {
+		rc = wait_event_interruptible_timeout(
+			link->wr_tx_wait,
+			!smc_link_sendable(link) ||
+			lgr->terminating ||
+			(smc_wr_tx_get_free_slot_index(link, &idx) != -EBUSY),
+			SMC_WR_TX_WAIT_FREE_SLOT_TIME);
+		if (!rc) {
+			/* timeout - terminate link */
+			smcr_link_down_cond_sched(link);
+			return -EPIPE;
+		}
+		if (idx == link->wr_tx_cnt)
+			return -EPIPE;
+	}
+	wr_id = smc_wr_tx_get_next_wr_id(link);
+	wr_pend = &link->wr_tx_pends[idx];
+	wr_pend->wr_id = wr_id;
+	wr_pend->handler = handler;
+	wr_pend->link = link;
+	wr_pend->idx = idx;
+	wr_ib = &link->wr_tx_ibs[idx];
+	wr_ib->wr_id = wr_id;
+	*wr_buf = &link->wr_tx_bufs[idx];
+	if (wr_rdma_buf)
+		*wr_rdma_buf = &link->wr_tx_rdmas[idx];
+	*wr_pend_priv = &wr_pend->priv;
+	return 0;
+}
+
+int smc_wr_tx_get_v2_slot(struct smc_link *link,
+			  smc_wr_tx_handler handler,
+			  struct smc_wr_v2_buf **wr_buf,
+			  struct smc_wr_tx_pend_priv **wr_pend_priv)
+{
+	struct smc_wr_tx_pend *wr_pend;
+	struct ib_send_wr *wr_ib;
+	u64 wr_id;
+
+	if (link->wr_tx_v2_pend->idx == link->wr_tx_cnt)
+		return -EBUSY;
+
+	*wr_buf = NULL;
+	*wr_pend_priv = NULL;
+	wr_id = smc_wr_tx_get_next_wr_id(link);
+	wr_pend = link->wr_tx_v2_pend;
+	wr_pend->wr_id = wr_id;
+	wr_pend->handler = handler;
+	wr_pend->link = link;
+	wr_pend->idx = link->wr_tx_cnt;
+	wr_ib = link->wr_tx_v2_ib;
+	wr_ib->wr_id = wr_id;
+	*wr_buf = link->lgr->wr_tx_buf_v2;
+	*wr_pend_priv = &wr_pend->priv;
+	return 0;
+}
+
+int smc_wr_tx_put_slot(struct smc_link *link,
+		       struct smc_wr_tx_pend_priv *wr_pend_priv)
+{
+	struct smc_wr_tx_pend *pend;
+
+	pend = container_of(wr_pend_priv, struct smc_wr_tx_pend, priv);
+	if (pend->idx < link->wr_tx_cnt) {
+		u32 idx = pend->idx;
+
+		/* clear the full struct smc_wr_tx_pend including .priv */
+		memset(&link->wr_tx_pends[idx], 0,
+		       sizeof(link->wr_tx_pends[idx]));
+		memset(&link->wr_tx_bufs[idx], 0,
+		       sizeof(link->wr_tx_bufs[idx]));
+		test_and_clear_bit(idx, link->wr_tx_mask);
+		wake_up(&link->wr_tx_wait);
+		return 1;
+	} else if (link->lgr->smc_version == SMC_V2 &&
+		   pend->idx == link->wr_tx_cnt) {
+		/* Large v2 buffer */
+		memset(&link->wr_tx_v2_pend, 0,
+		       sizeof(link->wr_tx_v2_pend));
+		memset(&link->lgr->wr_tx_buf_v2, 0,
+		       sizeof(link->lgr->wr_tx_buf_v2));
+		return 1;
+	}
+
+	return 0;
+}
+
+/* Send prepared WR slot via ib_post_send.
+ * @priv: pointer to smc_wr_tx_pend_priv identifying prepared message buffer
+ */
+int smc_wr_tx_send(struct smc_link *link, struct smc_wr_tx_pend_priv *priv)
+{
+	struct smc_wr_tx_pend *pend;
+	int rc;
+
+	ib_req_notify_cq(link->smcibdev->roce_cq_send,
+			 IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS);
+	pend = container_of(priv, struct smc_wr_tx_pend, priv);
+	rc = ib_post_send(link->roce_qp, &link->wr_tx_ibs[pend->idx], NULL);
+	if (rc) {
+		smc_wr_tx_put_slot(link, priv);
+		smcr_link_down_cond_sched(link);
+	}
+	return rc;
+}
+
+int smc_wr_tx_v2_send(struct smc_link *link, struct smc_wr_tx_pend_priv *priv,
+		      int len)
+{
+	int rc;
+
+	link->wr_tx_v2_ib->sg_list[0].length = len;
+	ib_req_notify_cq(link->smcibdev->roce_cq_send,
+			 IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS);
+	rc = ib_post_send(link->roce_qp, link->wr_tx_v2_ib, NULL);
+	if (rc) {
+		smc_wr_tx_put_slot(link, priv);
+		smcr_link_down_cond_sched(link);
+	}
+	return rc;
+}
+
+/* Send prepared WR slot via ib_post_send and wait for send completion
+ * notification.
+ * @priv: pointer to smc_wr_tx_pend_priv identifying prepared message buffer
+ */
+int smc_wr_tx_send_wait(struct smc_link *link, struct smc_wr_tx_pend_priv *priv,
+			unsigned long timeout)
+{
+	struct smc_wr_tx_pend *pend;
+	u32 pnd_idx;
+	int rc;
+
+	pend = container_of(priv, struct smc_wr_tx_pend, priv);
+	pend->compl_requested = 1;
+	pnd_idx = pend->idx;
+	init_completion(&link->wr_tx_compl[pnd_idx]);
+
+	rc = smc_wr_tx_send(link, priv);
+	if (rc)
+		return rc;
+	/* wait for completion by smc_wr_tx_process_cqe() */
+	rc = wait_for_completion_interruptible_timeout(
+					&link->wr_tx_compl[pnd_idx], timeout);
+	if (rc <= 0)
+		rc = -ENODATA;
+	if (rc > 0)
+		rc = 0;
+	return rc;
+}
+
+/* Register a memory region and wait for result. */
+int smc_wr_reg_send(struct smc_link *link, struct ib_mr *mr)
+{
+	int rc;
+
+	ib_req_notify_cq(link->smcibdev->roce_cq_send,
+			 IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS);
+	link->wr_reg_state = POSTED;
+	link->wr_reg.wr.wr_id = (u64)(uintptr_t)mr;
+	link->wr_reg.mr = mr;
+	link->wr_reg.key = mr->rkey;
+	rc = ib_post_send(link->roce_qp, &link->wr_reg.wr, NULL);
+	if (rc)
+		return rc;
+
+	percpu_ref_get(&link->wr_reg_refs);
+	rc = wait_event_interruptible_timeout(link->wr_reg_wait,
+					      (link->wr_reg_state != POSTED),
+					      SMC_WR_REG_MR_WAIT_TIME);
+	percpu_ref_put(&link->wr_reg_refs);
+	if (!rc) {
+		/* timeout - terminate link */
+		smcr_link_down_cond_sched(link);
+		return -EPIPE;
+	}
+	if (rc == -ERESTARTSYS)
+		return -EINTR;
+	switch (link->wr_reg_state) {
+	case CONFIRMED:
+		rc = 0;
+		break;
+	case FAILED:
+		rc = -EIO;
+		break;
+	case POSTED:
+		rc = -EPIPE;
+		break;
+	}
+	return rc;
+}
+
+/****************************** receive queue ********************************/
+
+int smc_wr_rx_register_handler(struct smc_wr_rx_handler *handler)
+{
+	struct smc_wr_rx_handler *h_iter;
+	int rc = 0;
+
+	spin_lock(&smc_wr_rx_hash_lock);
+	hash_for_each_possible(smc_wr_rx_hash, h_iter, list, handler->type) {
+		if (h_iter->type == handler->type) {
+			rc = -EEXIST;
+			goto out_unlock;
+		}
+	}
+	hash_add(smc_wr_rx_hash, &handler->list, handler->type);
+out_unlock:
+	spin_unlock(&smc_wr_rx_hash_lock);
+	return rc;
+}
+
+/* Demultiplex a received work request based on the message type to its handler.
+ * Relies on smc_wr_rx_hash having been completely filled before any IB WRs,
+ * and not being modified any more afterwards so we don't need to lock it.
+ */
+static inline void smc_wr_rx_demultiplex(struct ib_wc *wc)
+{
+	struct smc_link *link = (struct smc_link *)wc->qp->qp_context;
+	struct smc_wr_rx_handler *handler;
+	struct smc_wr_rx_hdr *wr_rx;
+	u64 temp_wr_id;
+	u32 index;
+
+	if (wc->byte_len < sizeof(*wr_rx))
+		return; /* short message */
+	temp_wr_id = wc->wr_id;
+	index = do_div(temp_wr_id, link->wr_rx_cnt);
+	wr_rx = (struct smc_wr_rx_hdr *)(link->wr_rx_bufs + index * link->wr_rx_buflen);
+	hash_for_each_possible(smc_wr_rx_hash, handler, list, wr_rx->type) {
+		if (handler->type == wr_rx->type)
+			handler->handler(wc, wr_rx);
+	}
+}
+
+static inline void smc_wr_rx_process_cqes(struct ib_wc wc[], int num)
+{
+	struct smc_link *link;
+	int i;
+
+	for (i = 0; i < num; i++) {
+		link = wc[i].qp->qp_context;
+		link->wr_rx_id_compl = wc[i].wr_id;
+		if (wc[i].status == IB_WC_SUCCESS) {
+			link->wr_rx_tstamp = jiffies;
+			smc_wr_rx_demultiplex(&wc[i]);
+			smc_wr_rx_post(link); /* refill WR RX */
+		} else {
+			/* handle status errors */
+			switch (wc[i].status) {
+			case IB_WC_RETRY_EXC_ERR:
+			case IB_WC_RNR_RETRY_EXC_ERR:
+			case IB_WC_WR_FLUSH_ERR:
+				smcr_link_down_cond_sched(link);
+				if (link->wr_rx_id_compl == link->wr_rx_id)
+					wake_up(&link->wr_rx_empty_wait);
+				break;
+			default:
+				smc_wr_rx_post(link); /* refill WR RX */
+				break;
+			}
+		}
+	}
+}
+
+static void smc_wr_rx_tasklet_fn(struct tasklet_struct *t)
+{
+	struct smc_ib_device *dev = from_tasklet(dev, t, recv_tasklet);
+	struct ib_wc wc[SMC_WR_MAX_POLL_CQE];
+	int polled = 0;
+	int rc;
+
+again:
+	polled++;
+	do {
+		memset(&wc, 0, sizeof(wc));
+		rc = ib_poll_cq(dev->roce_cq_recv, SMC_WR_MAX_POLL_CQE, wc);
+		if (polled == 1) {
+			ib_req_notify_cq(dev->roce_cq_recv,
+					 IB_CQ_SOLICITED_MASK
+					 | IB_CQ_REPORT_MISSED_EVENTS);
+		}
+		if (!rc)
+			break;
+		smc_wr_rx_process_cqes(&wc[0], rc);
+	} while (rc > 0);
+	if (polled == 1)
+		goto again;
+}
+
+void smc_wr_rx_cq_handler(struct ib_cq *ib_cq, void *cq_context)
+{
+	struct smc_ib_device *dev = (struct smc_ib_device *)cq_context;
+
+	tasklet_schedule(&dev->recv_tasklet);
+}
+
+int smc_wr_rx_post_init(struct smc_link *link)
+{
+	u32 i;
+	int rc = 0;
+
+	for (i = 0; i < link->wr_rx_cnt; i++)
+		rc = smc_wr_rx_post(link);
+	return rc;
+}
+
+/***************************** init, exit, misc ******************************/
+
+void smc_wr_remember_qp_attr(struct smc_link *lnk)
+{
+	struct ib_qp_attr *attr = &lnk->qp_attr;
+	struct ib_qp_init_attr init_attr;
+
+	memset(attr, 0, sizeof(*attr));
+	memset(&init_attr, 0, sizeof(init_attr));
+	ib_query_qp(lnk->roce_qp, attr,
+		    IB_QP_STATE |
+		    IB_QP_CUR_STATE |
+		    IB_QP_PKEY_INDEX |
+		    IB_QP_PORT |
+		    IB_QP_QKEY |
+		    IB_QP_AV |
+		    IB_QP_PATH_MTU |
+		    IB_QP_TIMEOUT |
+		    IB_QP_RETRY_CNT |
+		    IB_QP_RNR_RETRY |
+		    IB_QP_RQ_PSN |
+		    IB_QP_ALT_PATH |
+		    IB_QP_MIN_RNR_TIMER |
+		    IB_QP_SQ_PSN |
+		    IB_QP_PATH_MIG_STATE |
+		    IB_QP_CAP |
+		    IB_QP_DEST_QPN,
+		    &init_attr);
+
+	lnk->wr_tx_cnt = min_t(size_t, lnk->max_send_wr,
+			       lnk->qp_attr.cap.max_send_wr);
+	lnk->wr_rx_cnt = min_t(size_t, lnk->max_recv_wr,
+			       lnk->qp_attr.cap.max_recv_wr);
+}
+
+static void smc_wr_init_sge(struct smc_link *lnk)
+{
+	bool send_inline = (lnk->qp_attr.cap.max_inline_data > SMC_WR_TX_SIZE);
+	u32 i;
+
+	for (i = 0; i < lnk->wr_tx_cnt; i++) {
+		lnk->wr_tx_sges[i].addr = send_inline ? (uintptr_t)(&lnk->wr_tx_bufs[i]) :
+			lnk->wr_tx_dma_addr + i * SMC_WR_BUF_SIZE;
+		lnk->wr_tx_sges[i].length = SMC_WR_TX_SIZE;
+		lnk->wr_tx_sges[i].lkey = lnk->roce_pd->local_dma_lkey;
+		lnk->wr_tx_rdma_sges[i].tx_rdma_sge[0].wr_tx_rdma_sge[0].lkey =
+			lnk->roce_pd->local_dma_lkey;
+		lnk->wr_tx_rdma_sges[i].tx_rdma_sge[0].wr_tx_rdma_sge[1].lkey =
+			lnk->roce_pd->local_dma_lkey;
+		lnk->wr_tx_rdma_sges[i].tx_rdma_sge[1].wr_tx_rdma_sge[0].lkey =
+			lnk->roce_pd->local_dma_lkey;
+		lnk->wr_tx_rdma_sges[i].tx_rdma_sge[1].wr_tx_rdma_sge[1].lkey =
+			lnk->roce_pd->local_dma_lkey;
+		lnk->wr_tx_ibs[i].next = NULL;
+		lnk->wr_tx_ibs[i].sg_list = &lnk->wr_tx_sges[i];
+		lnk->wr_tx_ibs[i].num_sge = 1;
+		lnk->wr_tx_ibs[i].opcode = IB_WR_SEND;
+		lnk->wr_tx_ibs[i].send_flags =
+			IB_SEND_SIGNALED | IB_SEND_SOLICITED;
+		if (send_inline)
+			lnk->wr_tx_ibs[i].send_flags |= IB_SEND_INLINE;
+		lnk->wr_tx_rdmas[i].wr_tx_rdma[0].wr.opcode = IB_WR_RDMA_WRITE;
+		lnk->wr_tx_rdmas[i].wr_tx_rdma[1].wr.opcode = IB_WR_RDMA_WRITE;
+		lnk->wr_tx_rdmas[i].wr_tx_rdma[0].wr.sg_list =
+			lnk->wr_tx_rdma_sges[i].tx_rdma_sge[0].wr_tx_rdma_sge;
+		lnk->wr_tx_rdmas[i].wr_tx_rdma[1].wr.sg_list =
+			lnk->wr_tx_rdma_sges[i].tx_rdma_sge[1].wr_tx_rdma_sge;
+	}
+
+	if (lnk->lgr->smc_version == SMC_V2) {
+		lnk->wr_tx_v2_sge->addr = lnk->wr_tx_v2_dma_addr;
+		lnk->wr_tx_v2_sge->length = SMC_WR_BUF_V2_SIZE;
+		lnk->wr_tx_v2_sge->lkey = lnk->roce_pd->local_dma_lkey;
+
+		lnk->wr_tx_v2_ib->next = NULL;
+		lnk->wr_tx_v2_ib->sg_list = lnk->wr_tx_v2_sge;
+		lnk->wr_tx_v2_ib->num_sge = 1;
+		lnk->wr_tx_v2_ib->opcode = IB_WR_SEND;
+		lnk->wr_tx_v2_ib->send_flags =
+			IB_SEND_SIGNALED | IB_SEND_SOLICITED;
+	}
+
+	/* With SMC-Rv2 there can be messages larger than SMC_WR_TX_SIZE.
+	 * Each ib_recv_wr gets 2 sges, the second one is a spillover buffer
+	 * and the same buffer for all sges. When a larger message arrived then
+	 * the content of the first small sge is copied to the beginning of
+	 * the larger spillover buffer, allowing easy data mapping.
+	 */
+	for (i = 0; i < lnk->wr_rx_cnt; i++) {
+		int x = i * lnk->wr_rx_sge_cnt;
+
+		lnk->wr_rx_sges[x].addr =
+			lnk->wr_rx_dma_addr + i * lnk->wr_rx_buflen;
+		lnk->wr_rx_sges[x].length = smc_link_shared_v2_rxbuf(lnk) ?
+			SMC_WR_TX_SIZE : lnk->wr_rx_buflen;
+		lnk->wr_rx_sges[x].lkey = lnk->roce_pd->local_dma_lkey;
+		if (lnk->lgr->smc_version == SMC_V2 && smc_link_shared_v2_rxbuf(lnk)) {
+			lnk->wr_rx_sges[x + 1].addr =
+					lnk->wr_rx_v2_dma_addr + SMC_WR_TX_SIZE;
+			lnk->wr_rx_sges[x + 1].length =
+					SMC_WR_BUF_V2_SIZE - SMC_WR_TX_SIZE;
+			lnk->wr_rx_sges[x + 1].lkey =
+					lnk->roce_pd->local_dma_lkey;
+		}
+		lnk->wr_rx_ibs[i].next = NULL;
+		lnk->wr_rx_ibs[i].sg_list = &lnk->wr_rx_sges[x];
+		lnk->wr_rx_ibs[i].num_sge = lnk->wr_rx_sge_cnt;
+	}
+	lnk->wr_reg.wr.next = NULL;
+	lnk->wr_reg.wr.num_sge = 0;
+	lnk->wr_reg.wr.send_flags = IB_SEND_SIGNALED;
+	lnk->wr_reg.wr.opcode = IB_WR_REG_MR;
+	lnk->wr_reg.access = IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE;
+}
+
+void smc_wr_free_link(struct smc_link *lnk)
+{
+	struct ib_device *ibdev;
+
+	if (!lnk->smcibdev)
+		return;
+	ibdev = lnk->smcibdev->ibdev;
+
+	smc_wr_drain_cq(lnk);
+	smc_wr_wakeup_reg_wait(lnk);
+	smc_wr_wakeup_tx_wait(lnk);
+
+	smc_wr_tx_wait_no_pending_sends(lnk);
+	percpu_ref_kill(&lnk->wr_reg_refs);
+	wait_for_completion(&lnk->reg_ref_comp);
+	percpu_ref_exit(&lnk->wr_reg_refs);
+	percpu_ref_kill(&lnk->wr_tx_refs);
+	wait_for_completion(&lnk->tx_ref_comp);
+	percpu_ref_exit(&lnk->wr_tx_refs);
+
+	if (lnk->wr_rx_dma_addr) {
+		ib_dma_unmap_single(ibdev, lnk->wr_rx_dma_addr,
+				    lnk->wr_rx_buflen * lnk->wr_rx_cnt,
+				    DMA_FROM_DEVICE);
+		lnk->wr_rx_dma_addr = 0;
+	}
+	if (lnk->wr_rx_v2_dma_addr) {
+		ib_dma_unmap_single(ibdev, lnk->wr_rx_v2_dma_addr,
+				    SMC_WR_BUF_V2_SIZE,
+				    DMA_FROM_DEVICE);
+		lnk->wr_rx_v2_dma_addr = 0;
+	}
+	if (lnk->wr_tx_dma_addr) {
+		ib_dma_unmap_single(ibdev, lnk->wr_tx_dma_addr,
+				    SMC_WR_BUF_SIZE * lnk->wr_tx_cnt,
+				    DMA_TO_DEVICE);
+		lnk->wr_tx_dma_addr = 0;
+	}
+	if (lnk->wr_tx_v2_dma_addr) {
+		ib_dma_unmap_single(ibdev, lnk->wr_tx_v2_dma_addr,
+				    SMC_WR_BUF_V2_SIZE,
+				    DMA_TO_DEVICE);
+		lnk->wr_tx_v2_dma_addr = 0;
+	}
+}
+
+void smc_wr_free_lgr_mem(struct smc_link_group *lgr)
+{
+	if (lgr->smc_version < SMC_V2)
+		return;
+
+	kfree(lgr->wr_rx_buf_v2);
+	lgr->wr_rx_buf_v2 = NULL;
+	kfree(lgr->wr_tx_buf_v2);
+	lgr->wr_tx_buf_v2 = NULL;
+}
+
+void smc_wr_free_link_mem(struct smc_link *lnk)
+{
+	kfree(lnk->wr_tx_v2_ib);
+	lnk->wr_tx_v2_ib = NULL;
+	kfree(lnk->wr_tx_v2_sge);
+	lnk->wr_tx_v2_sge = NULL;
+	kfree(lnk->wr_tx_v2_pend);
+	lnk->wr_tx_v2_pend = NULL;
+	kfree(lnk->wr_tx_compl);
+	lnk->wr_tx_compl = NULL;
+	kfree(lnk->wr_tx_pends);
+	lnk->wr_tx_pends = NULL;
+	bitmap_free(lnk->wr_tx_mask);
+	lnk->wr_tx_mask = NULL;
+	kfree(lnk->wr_tx_sges);
+	lnk->wr_tx_sges = NULL;
+	kfree(lnk->wr_tx_rdma_sges);
+	lnk->wr_tx_rdma_sges = NULL;
+	kfree(lnk->wr_rx_sges);
+	lnk->wr_rx_sges = NULL;
+	kfree(lnk->wr_tx_rdmas);
+	lnk->wr_tx_rdmas = NULL;
+	kfree(lnk->wr_rx_ibs);
+	lnk->wr_rx_ibs = NULL;
+	kfree(lnk->wr_tx_ibs);
+	lnk->wr_tx_ibs = NULL;
+	kfree(lnk->wr_tx_bufs);
+	lnk->wr_tx_bufs = NULL;
+	kfree(lnk->wr_rx_bufs);
+	lnk->wr_rx_bufs = NULL;
+}
+
+int smc_wr_alloc_lgr_mem(struct smc_link_group *lgr)
+{
+	if (lgr->smc_version < SMC_V2)
+		return 0;
+
+	lgr->wr_rx_buf_v2 = kzalloc(SMC_WR_BUF_V2_SIZE, GFP_KERNEL);
+	if (!lgr->wr_rx_buf_v2)
+		return -ENOMEM;
+	lgr->wr_tx_buf_v2 = kzalloc(SMC_WR_BUF_V2_SIZE, GFP_KERNEL);
+	if (!lgr->wr_tx_buf_v2) {
+		kfree(lgr->wr_rx_buf_v2);
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+int smc_wr_alloc_link_mem(struct smc_link *link)
+{
+	/* allocate link related memory */
+	link->wr_tx_bufs = kcalloc(link->max_send_wr,
+				   SMC_WR_BUF_SIZE, GFP_KERNEL);
+	if (!link->wr_tx_bufs)
+		goto no_mem;
+	link->wr_rx_bufs = kcalloc(link->max_recv_wr, link->wr_rx_buflen,
+				   GFP_KERNEL);
+	if (!link->wr_rx_bufs)
+		goto no_mem_wr_tx_bufs;
+	link->wr_tx_ibs = kzalloc_objs(link->wr_tx_ibs[0], link->max_send_wr);
+	if (!link->wr_tx_ibs)
+		goto no_mem_wr_rx_bufs;
+	link->wr_rx_ibs = kzalloc_objs(link->wr_rx_ibs[0], link->max_recv_wr);
+	if (!link->wr_rx_ibs)
+		goto no_mem_wr_tx_ibs;
+	link->wr_tx_rdmas = kzalloc_objs(link->wr_tx_rdmas[0],
+					 link->max_send_wr);
+	if (!link->wr_tx_rdmas)
+		goto no_mem_wr_rx_ibs;
+	link->wr_tx_rdma_sges = kzalloc_objs(link->wr_tx_rdma_sges[0],
+					     link->max_send_wr);
+	if (!link->wr_tx_rdma_sges)
+		goto no_mem_wr_tx_rdmas;
+	link->wr_tx_sges = kzalloc_objs(link->wr_tx_sges[0], link->max_send_wr);
+	if (!link->wr_tx_sges)
+		goto no_mem_wr_tx_rdma_sges;
+	link->wr_rx_sges = kcalloc(link->max_recv_wr,
+				   sizeof(link->wr_rx_sges[0]) * link->wr_rx_sge_cnt,
+				   GFP_KERNEL);
+	if (!link->wr_rx_sges)
+		goto no_mem_wr_tx_sges;
+	link->wr_tx_mask = bitmap_zalloc(link->max_send_wr, GFP_KERNEL);
+	if (!link->wr_tx_mask)
+		goto no_mem_wr_rx_sges;
+	link->wr_tx_pends = kzalloc_objs(link->wr_tx_pends[0],
+					 link->max_send_wr);
+	if (!link->wr_tx_pends)
+		goto no_mem_wr_tx_mask;
+	link->wr_tx_compl = kzalloc_objs(link->wr_tx_compl[0],
+					 link->max_send_wr);
+	if (!link->wr_tx_compl)
+		goto no_mem_wr_tx_pends;
+
+	if (link->lgr->smc_version == SMC_V2) {
+		link->wr_tx_v2_ib = kzalloc_obj(*link->wr_tx_v2_ib);
+		if (!link->wr_tx_v2_ib)
+			goto no_mem_tx_compl;
+		link->wr_tx_v2_sge = kzalloc_obj(*link->wr_tx_v2_sge);
+		if (!link->wr_tx_v2_sge)
+			goto no_mem_v2_ib;
+		link->wr_tx_v2_pend = kzalloc_obj(*link->wr_tx_v2_pend);
+		if (!link->wr_tx_v2_pend)
+			goto no_mem_v2_sge;
+	}
+	return 0;
+
+no_mem_v2_sge:
+	kfree(link->wr_tx_v2_sge);
+no_mem_v2_ib:
+	kfree(link->wr_tx_v2_ib);
+no_mem_tx_compl:
+	kfree(link->wr_tx_compl);
+no_mem_wr_tx_pends:
+	kfree(link->wr_tx_pends);
+no_mem_wr_tx_mask:
+	kfree(link->wr_tx_mask);
+no_mem_wr_rx_sges:
+	kfree(link->wr_rx_sges);
+no_mem_wr_tx_sges:
+	kfree(link->wr_tx_sges);
+no_mem_wr_tx_rdma_sges:
+	kfree(link->wr_tx_rdma_sges);
+no_mem_wr_tx_rdmas:
+	kfree(link->wr_tx_rdmas);
+no_mem_wr_rx_ibs:
+	kfree(link->wr_rx_ibs);
+no_mem_wr_tx_ibs:
+	kfree(link->wr_tx_ibs);
+no_mem_wr_rx_bufs:
+	kfree(link->wr_rx_bufs);
+no_mem_wr_tx_bufs:
+	kfree(link->wr_tx_bufs);
+no_mem:
+	return -ENOMEM;
+}
+
+void smc_wr_remove_dev(struct smc_ib_device *smcibdev)
+{
+	tasklet_kill(&smcibdev->recv_tasklet);
+	tasklet_kill(&smcibdev->send_tasklet);
+}
+
+void smc_wr_add_dev(struct smc_ib_device *smcibdev)
+{
+	tasklet_setup(&smcibdev->recv_tasklet, smc_wr_rx_tasklet_fn);
+	tasklet_setup(&smcibdev->send_tasklet, smc_wr_tx_tasklet_fn);
+}
+
+static void smcr_wr_tx_refs_free(struct percpu_ref *ref)
+{
+	struct smc_link *lnk = container_of(ref, struct smc_link, wr_tx_refs);
+
+	complete(&lnk->tx_ref_comp);
+}
+
+static void smcr_wr_reg_refs_free(struct percpu_ref *ref)
+{
+	struct smc_link *lnk = container_of(ref, struct smc_link, wr_reg_refs);
+
+	complete(&lnk->reg_ref_comp);
+}
+
+int smc_wr_create_link(struct smc_link *lnk)
+{
+	struct ib_device *ibdev = lnk->smcibdev->ibdev;
+	int rc = 0;
+
+	smc_wr_tx_set_wr_id(&lnk->wr_tx_id, 0);
+	lnk->wr_rx_id = 0;
+	lnk->wr_rx_dma_addr = ib_dma_map_single(
+		ibdev, lnk->wr_rx_bufs,	lnk->wr_rx_buflen * lnk->wr_rx_cnt,
+		DMA_FROM_DEVICE);
+	if (ib_dma_mapping_error(ibdev, lnk->wr_rx_dma_addr)) {
+		lnk->wr_rx_dma_addr = 0;
+		rc = -EIO;
+		goto out;
+	}
+	if (lnk->lgr->smc_version == SMC_V2) {
+		if (smc_link_shared_v2_rxbuf(lnk)) {
+			lnk->wr_rx_v2_dma_addr =
+				ib_dma_map_single(ibdev, lnk->lgr->wr_rx_buf_v2,
+						  SMC_WR_BUF_V2_SIZE, DMA_FROM_DEVICE);
+			if (ib_dma_mapping_error(ibdev, lnk->wr_rx_v2_dma_addr)) {
+				lnk->wr_rx_v2_dma_addr = 0;
+				rc = -EIO;
+				goto dma_unmap;
+			}
+		}
+		lnk->wr_tx_v2_dma_addr = ib_dma_map_single(ibdev,
+			lnk->lgr->wr_tx_buf_v2, SMC_WR_BUF_V2_SIZE,
+			DMA_TO_DEVICE);
+		if (ib_dma_mapping_error(ibdev, lnk->wr_tx_v2_dma_addr)) {
+			lnk->wr_tx_v2_dma_addr = 0;
+			rc = -EIO;
+			goto dma_unmap;
+		}
+	}
+	lnk->wr_tx_dma_addr = ib_dma_map_single(
+		ibdev, lnk->wr_tx_bufs,	SMC_WR_BUF_SIZE * lnk->wr_tx_cnt,
+		DMA_TO_DEVICE);
+	if (ib_dma_mapping_error(ibdev, lnk->wr_tx_dma_addr)) {
+		rc = -EIO;
+		goto dma_unmap;
+	}
+	smc_wr_init_sge(lnk);
+	bitmap_zero(lnk->wr_tx_mask, lnk->max_send_wr);
+	init_waitqueue_head(&lnk->wr_tx_wait);
+	rc = percpu_ref_init(&lnk->wr_tx_refs, smcr_wr_tx_refs_free, 0, GFP_KERNEL);
+	if (rc)
+		goto dma_unmap;
+	init_completion(&lnk->tx_ref_comp);
+	init_waitqueue_head(&lnk->wr_reg_wait);
+	rc = percpu_ref_init(&lnk->wr_reg_refs, smcr_wr_reg_refs_free, 0, GFP_KERNEL);
+	if (rc)
+		goto cancel_ref;
+	init_completion(&lnk->reg_ref_comp);
+	init_waitqueue_head(&lnk->wr_rx_empty_wait);
+	return rc;
+
+cancel_ref:
+	percpu_ref_exit(&lnk->wr_tx_refs);
+dma_unmap:
+	if (lnk->wr_rx_v2_dma_addr) {
+		ib_dma_unmap_single(ibdev, lnk->wr_rx_v2_dma_addr,
+				    SMC_WR_BUF_V2_SIZE,
+				    DMA_FROM_DEVICE);
+		lnk->wr_rx_v2_dma_addr = 0;
+	}
+	if (lnk->wr_tx_v2_dma_addr) {
+		ib_dma_unmap_single(ibdev, lnk->wr_tx_v2_dma_addr,
+				    SMC_WR_BUF_V2_SIZE,
+				    DMA_TO_DEVICE);
+		lnk->wr_tx_v2_dma_addr = 0;
+	}
+	ib_dma_unmap_single(ibdev, lnk->wr_rx_dma_addr,
+			    lnk->wr_rx_buflen * lnk->wr_rx_cnt,
+			    DMA_FROM_DEVICE);
+	lnk->wr_rx_dma_addr = 0;
+out:
+	return rc;
+}

linux-7.2.3/net/smc/smc_wr.h

@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ * Work Requests exploiting Infiniband API
+ *
+ * Copyright IBM Corp. 2016
+ *
+ * Author(s):  Steffen Maier <maier@linux.vnet.ibm.com>
+ */
+
+#ifndef SMC_WR_H
+#define SMC_WR_H
+
+#include <linux/atomic.h>
+#include <rdma/ib_verbs.h>
+#include <asm/div64.h>
+
+#include "smc.h"
+#include "smc_core.h"
+
+#define SMC_WR_TX_WAIT_FREE_SLOT_TIME	(10 * HZ)
+
+#define SMC_WR_TX_SIZE 44 /* actual size of wr_send data (<=SMC_WR_BUF_SIZE) */
+
+#define SMC_WR_TX_PEND_PRIV_SIZE 32
+
+struct smc_wr_tx_pend_priv {
+	u8			priv[SMC_WR_TX_PEND_PRIV_SIZE];
+};
+
+typedef void (*smc_wr_tx_handler)(struct smc_wr_tx_pend_priv *,
+				  struct smc_link *,
+				  enum ib_wc_status);
+
+typedef bool (*smc_wr_tx_filter)(struct smc_wr_tx_pend_priv *,
+				 unsigned long);
+
+typedef void (*smc_wr_tx_dismisser)(struct smc_wr_tx_pend_priv *);
+
+struct smc_wr_rx_handler {
+	struct hlist_node	list;	/* hash table collision resolution */
+	void			(*handler)(struct ib_wc *, void *);
+	u8			type;
+};
+
+/* Only used by RDMA write WRs.
+ * All other WRs (CDC/LLC) use smc_wr_tx_send handling WR_ID implicitly
+ */
+static inline long smc_wr_tx_get_next_wr_id(struct smc_link *link)
+{
+	return atomic_long_inc_return(&link->wr_tx_id);
+}
+
+static inline void smc_wr_tx_set_wr_id(atomic_long_t *wr_tx_id, long val)
+{
+	atomic_long_set(wr_tx_id, val);
+}
+
+static inline bool smc_wr_tx_link_hold(struct smc_link *link)
+{
+	if (!smc_link_sendable(link))
+		return false;
+	percpu_ref_get(&link->wr_tx_refs);
+	return true;
+}
+
+static inline void smc_wr_tx_link_put(struct smc_link *link)
+{
+	percpu_ref_put(&link->wr_tx_refs);
+}
+
+static inline void smc_wr_drain_cq(struct smc_link *lnk)
+{
+	wait_event(lnk->wr_rx_empty_wait, lnk->wr_rx_id_compl == lnk->wr_rx_id);
+}
+
+static inline void smc_wr_wakeup_tx_wait(struct smc_link *lnk)
+{
+	wake_up_all(&lnk->wr_tx_wait);
+}
+
+static inline void smc_wr_wakeup_reg_wait(struct smc_link *lnk)
+{
+	wake_up(&lnk->wr_reg_wait);
+}
+
+/* post a new receive work request to fill a completed old work request entry */
+static inline int smc_wr_rx_post(struct smc_link *link)
+{
+	int rc;
+	u64 wr_id, temp_wr_id;
+	u32 index;
+
+	wr_id = ++link->wr_rx_id; /* tasklet context, thus not atomic */
+	temp_wr_id = wr_id;
+	index = do_div(temp_wr_id, link->wr_rx_cnt);
+	link->wr_rx_ibs[index].wr_id = wr_id;
+	rc = ib_post_recv(link->roce_qp, &link->wr_rx_ibs[index], NULL);
+	return rc;
+}
+
+int smc_wr_create_link(struct smc_link *lnk);
+int smc_wr_alloc_link_mem(struct smc_link *lnk);
+int smc_wr_alloc_lgr_mem(struct smc_link_group *lgr);
+void smc_wr_free_link(struct smc_link *lnk);
+void smc_wr_free_link_mem(struct smc_link *lnk);
+void smc_wr_free_lgr_mem(struct smc_link_group *lgr);
+void smc_wr_remember_qp_attr(struct smc_link *lnk);
+void smc_wr_remove_dev(struct smc_ib_device *smcibdev);
+void smc_wr_add_dev(struct smc_ib_device *smcibdev);
+
+int smc_wr_tx_get_free_slot(struct smc_link *link, smc_wr_tx_handler handler,
+			    struct smc_wr_buf **wr_buf,
+			    struct smc_rdma_wr **wrs,
+			    struct smc_wr_tx_pend_priv **wr_pend_priv);
+int smc_wr_tx_get_v2_slot(struct smc_link *link,
+			  smc_wr_tx_handler handler,
+			  struct smc_wr_v2_buf **wr_buf,
+			  struct smc_wr_tx_pend_priv **wr_pend_priv);
+int smc_wr_tx_put_slot(struct smc_link *link,
+		       struct smc_wr_tx_pend_priv *wr_pend_priv);
+int smc_wr_tx_send(struct smc_link *link,
+		   struct smc_wr_tx_pend_priv *wr_pend_priv);
+int smc_wr_tx_v2_send(struct smc_link *link,
+		      struct smc_wr_tx_pend_priv *priv, int len);
+int smc_wr_tx_send_wait(struct smc_link *link, struct smc_wr_tx_pend_priv *priv,
+			unsigned long timeout);
+void smc_wr_tx_cq_handler(struct ib_cq *ib_cq, void *cq_context);
+void smc_wr_tx_wait_no_pending_sends(struct smc_link *link);
+
+int smc_wr_rx_register_handler(struct smc_wr_rx_handler *handler);
+int smc_wr_rx_post_init(struct smc_link *link);
+void smc_wr_rx_cq_handler(struct ib_cq *ib_cq, void *cq_context);
+int smc_wr_reg_send(struct smc_link *link, struct ib_mr *mr);
+
+#endif /* SMC_WR_H */