[linux-2.6-block.git] / net / smc / smc_tx.c

// 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_tx.h"

#define SMC_TX_WORK_DELAY	HZ
#define SMC_TX_CORK_DELAY	(HZ >> 2)	/* 250 ms */

/***************************** sndbuf producer *******************************/

/* callback implementation for sk.sk_write_space()
 * to wakeup sndbuf producers that blocked with smc_tx_wait_memory().
 * 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) {
		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_memory().
 * 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 */
static int smc_tx_wait_memory(struct smc_sock *smc, int flags)
{
	DEFINE_WAIT_FUNC(wait, woken_wake_function);
	struct smc_connection *conn = &smc->conn;
	struct sock *sk = &smc->sk;
	bool noblock;
	long timeo;
	int rc = 0;

	/* similar to sk_stream_wait_memory */
	timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
	noblock = timeo ? false : true;
	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->local_tx_ctrl.conn_state_flags.peer_done_writing) {
			rc = -EPIPE;
			break;
		}
		if (smc_cdc_rxed_any_close(conn)) {
			rc = -ECONNRESET;
			break;
		}
		if (!timeo) {
			if (noblock)
				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))
			break; /* at least 1 byte of free space available */
		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
		sk_wait_event(sk, &timeo,
			      sk->sk_err ||
			      (sk->sk_shutdown & SEND_SHUTDOWN) ||
			      smc_cdc_rxed_any_close(conn) ||
			      atomic_read(&conn->sndbuf_space),
			      &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;
}

/* 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;
	}

	while (msg_data_left(msg)) {
		if (sk->sk_state == SMC_INIT)
			return -ENOTCONN;
		if (smc->sk.sk_shutdown & SEND_SHUTDOWN ||
		    (smc->sk.sk_err == ECONNABORTED) ||
		    conn->local_tx_ctrl.conn_state_flags.peer_conn_abort)
			return -EPIPE;
		if (smc_cdc_rxed_any_close(conn))
			return send_done ?: -ECONNRESET;

		if (!atomic_read(&conn->sndbuf_space)) {
			rc = smc_tx_wait_memory(smc, msg->msg_flags);
			if (rc) {
				if (send_done)
					return send_done;
				goto out_err;
			}
			continue;
		}

		/* initialize variables for 1st iteration of subsequent loop */
		/* could be just 1 byte, even after smc_tx_wait_memory 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_write(&prep,
			       smc_curs_read(&conn->tx_curs_prep, conn),
			       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;
		smc_sndbuf_sync_sg_for_cpu(conn);
		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_write(&conn->tx_curs_prep,
			       smc_curs_read(&prep, conn),
			       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_MORE || smc_tx_is_corked(smc)) &&
		    (atomic_read(&conn->sndbuf_space) >
						(conn->sndbuf_desc->len >> 1)))
			/* for a corked socket defer the RDMA writes if there
			 * is still sufficient sndbuf_space available
			 */
			schedule_delayed_work(&conn->tx_work,
					      SMC_TX_CORK_DELAY);
		else
			smc_tx_sndbuf_nonempty(conn);
	} /* 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 RDMA write */
static int smc_tx_rdma_write(struct smc_connection *conn, int peer_rmbe_offset,
			     int num_sges, struct ib_sge sges[])
{
	struct smc_link_group *lgr = conn->lgr;
	struct ib_send_wr *failed_wr = NULL;
	struct ib_rdma_wr rdma_wr;
	struct smc_link *link;
	int rc;

	memset(&rdma_wr, 0, sizeof(rdma_wr));
	link = &lgr->lnk[SMC_SINGLE_LINK];
	rdma_wr.wr.wr_id = smc_wr_tx_get_next_wr_id(link);
	rdma_wr.wr.sg_list = sges;
	rdma_wr.wr.num_sge = num_sges;
	rdma_wr.wr.opcode = IB_WR_RDMA_WRITE;
	rdma_wr.remote_addr =
		lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].dma_addr +
		/* RMBE within RMB */
		((conn->peer_rmbe_idx - 1) * conn->peer_rmbe_size) +
		/* offset within RMBE */
		peer_rmbe_offset;
	rdma_wr.rkey = lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].rkey;
	rc = ib_post_send(link->roce_qp, &rdma_wr.wr, &failed_wr);
	if (rc) {
		conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
		smc_lgr_terminate(lgr);
	}
	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);
}

/* 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)
{
	size_t src_off, src_len, dst_off, dst_len; /* current chunk values */
	size_t len, dst_len_sum, src_len_sum, dstchunk, srcchunk;
	union smc_host_cursor sent, prep, prod, cons;
	struct ib_sge sges[SMC_IB_MAX_SEND_SGE];
	struct smc_link_group *lgr = conn->lgr;
	int to_send, rmbespace;
	struct smc_link *link;
	dma_addr_t dma_addr;
	int num_sges;
	int rc;

	/* source: sndbuf */
	smc_curs_write(&sent, smc_curs_read(&conn->tx_curs_sent, conn), conn);
	smc_curs_write(&prep, smc_curs_read(&conn->tx_curs_prep, conn), 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)
		return 0;
	smc_curs_write(&prod,
		       smc_curs_read(&conn->local_tx_ctrl.prod, conn),
		       conn);
	smc_curs_write(&cons,
		       smc_curs_read(&conn->local_rx_ctrl.cons, conn),
		       conn);

	/* if usable snd_wnd closes ask peer to advertise once it opens again */
	conn->local_tx_ctrl.prod_flags.write_blocked = (to_send >= rmbespace);
	/* cf. usable snd_wnd */
	len = min(to_send, rmbespace);

	/* initialize variables for first iteration of subsequent nested loop */
	link = &lgr->lnk[SMC_SINGLE_LINK];
	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_sum = dst_len;
	src_off = sent.count;
	/* 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;
	}
	src_len_sum = src_len;
	dma_addr = sg_dma_address(conn->sndbuf_desc->sgt[SMC_SINGLE_LINK].sgl);
	for (dstchunk = 0; dstchunk < 2; dstchunk++) {
		num_sges = 0;
		for (srcchunk = 0; srcchunk < 2; srcchunk++) {
			sges[srcchunk].addr = dma_addr + src_off;
			sges[srcchunk].length = src_len;
			sges[srcchunk].lkey = link->roce_pd->local_dma_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, sges);
		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;
	}

	smc_tx_advance_cursors(conn, &prod, &sent, len);
	/* update connection's cursors with advanced local cursors */
	smc_curs_write(&conn->local_tx_ctrl.prod,
		       smc_curs_read(&prod, conn),
		       conn);
							/* dst: peer RMBE */
	smc_curs_write(&conn->tx_curs_sent,
		       smc_curs_read(&sent, conn),
		       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
 */
int smc_tx_sndbuf_nonempty(struct smc_connection *conn)
{
	struct smc_cdc_tx_pend *pend;
	struct smc_wr_buf *wr_buf;
	int rc;

	spin_lock_bh(&conn->send_lock);
	rc = smc_cdc_get_free_slot(conn, &wr_buf, &pend);
	if (rc < 0) {
		if (rc == -EBUSY) {
			struct smc_sock *smc =
				container_of(conn, struct smc_sock, conn);

			if (smc->sk.sk_err == ECONNABORTED) {
				rc = sock_error(&smc->sk);
				goto out_unlock;
			}
			rc = 0;
			if (conn->alert_token_local) /* connection healthy */
				mod_delayed_work(system_wq, &conn->tx_work,
						 SMC_TX_WORK_DELAY);
		}
		goto out_unlock;
	}

	rc = smc_tx_rdma_writes(conn);
	if (rc) {
		smc_wr_tx_put_slot(&conn->lgr->lnk[SMC_SINGLE_LINK],
				   (struct smc_wr_tx_pend_priv *)pend);
		goto out_unlock;
	}

	rc = smc_cdc_msg_send(conn, wr_buf, pend);

out_unlock:
	spin_unlock_bh(&conn->send_lock);
	return rc;
}

/* Wakeup sndbuf consumers from process context
 * since there is more data to transmit
 */
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);
	int rc;

	lock_sock(&smc->sk);
	if (smc->sk.sk_err ||
	    !conn->alert_token_local ||
	    conn->local_rx_ctrl.conn_state_flags.peer_conn_abort)
		goto out;

	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;

out:
	release_sock(&smc->sk);
}

void smc_tx_consumer_update(struct smc_connection *conn)
{
	union smc_host_cursor cfed, cons;
	int to_confirm;

	smc_curs_write(&cons,
		       smc_curs_read(&conn->local_tx_ctrl.cons, conn),
		       conn);
	smc_curs_write(&cfed,
		       smc_curs_read(&conn->rx_curs_confirmed, conn),
		       conn);
	to_confirm = smc_curs_diff(conn->rmb_desc->len, &cfed, &cons);

	if (conn->local_rx_ctrl.prod_flags.cons_curs_upd_req ||
	    ((to_confirm > conn->rmbe_update_limit) &&
	     ((to_confirm > (conn->rmb_desc->len / 2)) ||
	      conn->local_rx_ctrl.prod_flags.write_blocked))) {
		if ((smc_cdc_get_slot_and_msg_send(conn) < 0) &&
		    conn->alert_token_local) { /* connection healthy */
			schedule_delayed_work(&conn->tx_work,
					      SMC_TX_WORK_DELAY);
			return;
		}
		smc_curs_write(&conn->rx_curs_confirmed,
			       smc_curs_read(&conn->local_tx_ctrl.cons, conn),
			       conn);
		conn->local_rx_ctrl.prod_flags.cons_curs_upd_req = 0;
	}
	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;
}
Commit	Line	Data
b2441318	1	// SPDX-License-Identifier: GPL-2.0
e6727f39 UB	2	/*
	3	* Shared Memory Communications over RDMA (SMC-R) and RoCE
	4	*
	5	* Manage send buffer.
	6	* Producer:
	7	* Copy user space data into send buffer, if send buffer space available.
	8	* Consumer:
	9	* Trigger RDMA write into RMBE of peer and send CDC, if RMBE space available.
	10	*
	11	* Copyright IBM Corp. 2016
	12	*
	13	* Author(s): Ursula Braun <ubraun@linux.vnet.ibm.com>
	14	*/
	15
	16	#include <linux/net.h>
	17	#include <linux/rcupdate.h>
	18	#include <linux/workqueue.h>
c3edc401 IM	19	#include <linux/sched/signal.h>
c3edc401 IM	20
e6727f39	21	#include <net/sock.h>
01d2f7e2	22	#include <net/tcp.h>
e6727f39 UB	23
	24	#include "smc.h"
	25	#include "smc_wr.h"
	26	#include "smc_cdc.h"
	27	#include "smc_tx.h"
	28
18e537cd	29	#define SMC_TX_WORK_DELAY HZ
01d2f7e2	30	#define SMC_TX_CORK_DELAY (HZ >> 2) /* 250 ms */
18e537cd	31
e6727f39 UB	32	/*************************** sndbuf producer *****************************/
	33
	34	/* callback implementation for sk.sk_write_space()
	35	* to wakeup sndbuf producers that blocked with smc_tx_wait_memory().
	36	* called under sk_socket lock.
	37	*/
	38	static void smc_tx_write_space(struct sock *sk)
	39	{
	40	struct socket *sock = sk->sk_socket;
	41	struct smc_sock *smc = smc_sk(sk);
	42	struct socket_wq *wq;
	43
	44	/* similar to sk_stream_write_space */
	45	if (atomic_read(&smc->conn.sndbuf_space) && sock) {
	46	clear_bit(SOCK_NOSPACE, &sock->flags);
	47	rcu_read_lock();
	48	wq = rcu_dereference(sk->sk_wq);
	49	if (skwq_has_sleeper(wq))
	50	wake_up_interruptible_poll(&wq->wait,
a9a08845 LT	51	EPOLLOUT \| EPOLLWRNORM \|
a9a08845 LT	52	EPOLLWRBAND);
e6727f39 UB	53	if (wq && wq->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN))
	54	sock_wake_async(wq, SOCK_WAKE_SPACE, POLL_OUT);
	55	rcu_read_unlock();
	56	}
	57	}
	58
	59	/* Wakeup sndbuf producers that blocked with smc_tx_wait_memory().
	60	* Cf. tcp_data_snd_check()=>tcp_check_space()=>tcp_new_space().
	61	*/
	62	void smc_tx_sndbuf_nonfull(struct smc_sock *smc)
	63	{
	64	if (smc->sk.sk_socket &&
	65	test_bit(SOCK_NOSPACE, &smc->sk.sk_socket->flags))
	66	smc->sk.sk_write_space(&smc->sk);
	67	}
	68
	69	/* blocks sndbuf producer until at least one byte of free space available */
	70	static int smc_tx_wait_memory(struct smc_sock *smc, int flags)
	71	{
	72	DEFINE_WAIT_FUNC(wait, woken_wake_function);
	73	struct smc_connection *conn = &smc->conn;
	74	struct sock *sk = &smc->sk;
	75	bool noblock;
	76	long timeo;
	77	int rc = 0;
	78
	79	/* similar to sk_stream_wait_memory */
	80	timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
	81	noblock = timeo ? false : true;
	82	add_wait_queue(sk_sleep(sk), &wait);
	83	while (1) {
	84	sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
	85	if (sk->sk_err \|\|
	86	(sk->sk_shutdown & SEND_SHUTDOWN) \|\|
	87	conn->local_tx_ctrl.conn_state_flags.peer_done_writing) {
	88	rc = -EPIPE;
	89	break;
	90	}
aa377e68	91	if (smc_cdc_rxed_any_close(conn)) {
e6727f39 UB	92	rc = -ECONNRESET;
	93	break;
	94	}
	95	if (!timeo) {
	96	if (noblock)
	97	set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
	98	rc = -EAGAIN;
	99	break;
	100	}
	101	if (signal_pending(current)) {
	102	rc = sock_intr_errno(timeo);
	103	break;
	104	}
	105	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
	106	if (atomic_read(&conn->sndbuf_space))
	107	break; /* at least 1 byte of free space available */
	108	set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
e6727f39 UB	109	sk_wait_event(sk, &timeo,
	110	sk->sk_err \|\|
	111	(sk->sk_shutdown & SEND_SHUTDOWN) \|\|
aa377e68	112	smc_cdc_rxed_any_close(conn) \|\|
e6727f39 UB	113	atomic_read(&conn->sndbuf_space),
e6727f39 UB	114	&wait);
e6727f39 UB	115	}
	116	remove_wait_queue(sk_sleep(sk), &wait);
	117	return rc;
	118	}
	119
01d2f7e2 UB	120	static bool smc_tx_is_corked(struct smc_sock *smc)
	121	{
	122	struct tcp_sock *tp = tcp_sk(smc->clcsock->sk);
	123
	124	return (tp->nonagle & TCP_NAGLE_CORK) ? true : false;
	125	}
	126
e6727f39 UB	127	/* sndbuf producer: main API called by socket layer.
	128	* called under sock lock.
	129	*/
	130	int smc_tx_sendmsg(struct smc_sock smc, struct msghdr msg, size_t len)
	131	{
	132	size_t copylen, send_done = 0, send_remaining = len;
	133	size_t chunk_len, chunk_off, chunk_len_sum;
	134	struct smc_connection *conn = &smc->conn;
	135	union smc_host_cursor prep;
	136	struct sock *sk = &smc->sk;
	137	char *sndbuf_base;
	138	int tx_cnt_prep;
	139	int writespace;
	140	int rc, chunk;
	141
	142	/* This should be in poll */
	143	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
	144
	145	if (sk->sk_err \|\| (sk->sk_shutdown & SEND_SHUTDOWN)) {
	146	rc = -EPIPE;
	147	goto out_err;
	148	}
	149
	150	while (msg_data_left(msg)) {
	151	if (sk->sk_state == SMC_INIT)
	152	return -ENOTCONN;
	153	if (smc->sk.sk_shutdown & SEND_SHUTDOWN \|\|
b38d7324	154	(smc->sk.sk_err == ECONNABORTED) \|\|
e6727f39 UB	155	conn->local_tx_ctrl.conn_state_flags.peer_conn_abort)
	156	return -EPIPE;
	157	if (smc_cdc_rxed_any_close(conn))
	158	return send_done ?: -ECONNRESET;
	159
	160	if (!atomic_read(&conn->sndbuf_space)) {
	161	rc = smc_tx_wait_memory(smc, msg->msg_flags);
	162	if (rc) {
	163	if (send_done)
	164	return send_done;
	165	goto out_err;
	166	}
	167	continue;
	168	}
	169
	170	/* initialize variables for 1st iteration of subsequent loop */
	171	/* could be just 1 byte, even after smc_tx_wait_memory above */
	172	writespace = atomic_read(&conn->sndbuf_space);
	173	/* not more than what user space asked for */
	174	copylen = min_t(size_t, send_remaining, writespace);
	175	/* determine start of sndbuf */
	176	sndbuf_base = conn->sndbuf_desc->cpu_addr;
	177	smc_curs_write(&prep,
	178	smc_curs_read(&conn->tx_curs_prep, conn),
	179	conn);
	180	tx_cnt_prep = prep.count;
	181	/* determine chunks where to write into sndbuf */
	182	/* either unwrapped case, or 1st chunk of wrapped case */
69cb7dc0 HW	183	chunk_len = min_t(size_t, copylen, conn->sndbuf_desc->len -
69cb7dc0 HW	184	tx_cnt_prep);
e6727f39 UB	185	chunk_len_sum = chunk_len;
e6727f39 UB	186	chunk_off = tx_cnt_prep;
10428dd8	187	smc_sndbuf_sync_sg_for_cpu(conn);
e6727f39 UB	188	for (chunk = 0; chunk < 2; chunk++) {
	189	rc = memcpy_from_msg(sndbuf_base + chunk_off,
	190	msg, chunk_len);
	191	if (rc) {
10428dd8	192	smc_sndbuf_sync_sg_for_device(conn);
e6727f39 UB	193	if (send_done)
	194	return send_done;
	195	goto out_err;
	196	}
	197	send_done += chunk_len;
	198	send_remaining -= chunk_len;
	199
	200	if (chunk_len_sum == copylen)
	201	break; /* either on 1st or 2nd iteration */
	202	/* prepare next (== 2nd) iteration */
	203	chunk_len = copylen - chunk_len; /* remainder */
	204	chunk_len_sum += chunk_len;
	205	chunk_off = 0; /* modulo offset in send ring buffer */
	206	}
10428dd8	207	smc_sndbuf_sync_sg_for_device(conn);
e6727f39	208	/* update cursors */
69cb7dc0	209	smc_curs_add(conn->sndbuf_desc->len, &prep, copylen);
e6727f39 UB	210	smc_curs_write(&conn->tx_curs_prep,
	211	smc_curs_read(&prep, conn),
	212	conn);
	213	/* increased in send tasklet smc_cdc_tx_handler() */
	214	smp_mb__before_atomic();
	215	atomic_sub(copylen, &conn->sndbuf_space);
69cb7dc0	216	/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
e6727f39 UB	217	smp_mb__after_atomic();
	218	/* since we just produced more new data into sndbuf,
	219	* trigger sndbuf consumer: RDMA write into peer RMBE and CDC
	220	*/
01d2f7e2 UB	221	if ((msg->msg_flags & MSG_MORE \|\| smc_tx_is_corked(smc)) &&
01d2f7e2 UB	222	(atomic_read(&conn->sndbuf_space) >
69cb7dc0	223	(conn->sndbuf_desc->len >> 1)))
01d2f7e2 UB	224	/* for a corked socket defer the RDMA writes if there
	225	* is still sufficient sndbuf_space available
	226	*/
	227	schedule_delayed_work(&conn->tx_work,
	228	SMC_TX_CORK_DELAY);
	229	else
	230	smc_tx_sndbuf_nonempty(conn);
e6727f39 UB	231	} /* while (msg_data_left(msg)) */
	232
	233	return send_done;
	234
	235	out_err:
	236	rc = sk_stream_error(sk, msg->msg_flags, rc);
	237	/* make sure we wake any epoll edge trigger waiter */
	238	if (unlikely(rc == -EAGAIN))
	239	sk->sk_write_space(sk);
	240	return rc;
	241	}
	242
	243	/*************************** sndbuf consumer *****************************/
	244
	245	/* sndbuf consumer: actual data transfer of one target chunk with RDMA write */
	246	static int smc_tx_rdma_write(struct smc_connection *conn, int peer_rmbe_offset,
	247	int num_sges, struct ib_sge sges[])
	248	{
	249	struct smc_link_group *lgr = conn->lgr;
	250	struct ib_send_wr *failed_wr = NULL;
	251	struct ib_rdma_wr rdma_wr;
	252	struct smc_link *link;
	253	int rc;
	254
	255	memset(&rdma_wr, 0, sizeof(rdma_wr));
	256	link = &lgr->lnk[SMC_SINGLE_LINK];
	257	rdma_wr.wr.wr_id = smc_wr_tx_get_next_wr_id(link);
	258	rdma_wr.wr.sg_list = sges;
	259	rdma_wr.wr.num_sge = num_sges;
	260	rdma_wr.wr.opcode = IB_WR_RDMA_WRITE;
	261	rdma_wr.remote_addr =
	262	lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].dma_addr +
	263	/* RMBE within RMB */
92a138e3	264	((conn->peer_rmbe_idx - 1) * conn->peer_rmbe_size) +
e6727f39 UB	265	/* offset within RMBE */
	266	peer_rmbe_offset;
	267	rdma_wr.rkey = lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].rkey;
	268	rc = ib_post_send(link->roce_qp, &rdma_wr.wr, &failed_wr);
b4772b3a	269	if (rc) {
e6727f39	270	conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1;
b4772b3a UB	271	smc_lgr_terminate(lgr);
b4772b3a UB	272	}
e6727f39 UB	273	return rc;
	274	}
	275
	276	/* sndbuf consumer */
	277	static inline void smc_tx_advance_cursors(struct smc_connection *conn,
	278	union smc_host_cursor *prod,
	279	union smc_host_cursor *sent,
	280	size_t len)
	281	{
	282	smc_curs_add(conn->peer_rmbe_size, prod, len);
	283	/* increased in recv tasklet smc_cdc_msg_rcv() */
	284	smp_mb__before_atomic();
	285	/* data in flight reduces usable snd_wnd */
	286	atomic_sub(len, &conn->peer_rmbe_space);
	287	/* guarantee 0 <= peer_rmbe_space <= peer_rmbe_size */
	288	smp_mb__after_atomic();
69cb7dc0	289	smc_curs_add(conn->sndbuf_desc->len, sent, len);
e6727f39 UB	290	}
	291
	292	/* sndbuf consumer: prepare all necessary (src&dst) chunks of data transmit;
	293	* usable snd_wnd as max transmit
	294	*/
	295	static int smc_tx_rdma_writes(struct smc_connection *conn)
	296	{
	297	size_t src_off, src_len, dst_off, dst_len; /* current chunk values */
	298	size_t len, dst_len_sum, src_len_sum, dstchunk, srcchunk;
	299	union smc_host_cursor sent, prep, prod, cons;
	300	struct ib_sge sges[SMC_IB_MAX_SEND_SGE];
	301	struct smc_link_group *lgr = conn->lgr;
	302	int to_send, rmbespace;
	303	struct smc_link *link;
9d8fb617	304	dma_addr_t dma_addr;
e6727f39 UB	305	int num_sges;
	306	int rc;
	307
	308	/* source: sndbuf */
	309	smc_curs_write(&sent, smc_curs_read(&conn->tx_curs_sent, conn), conn);
	310	smc_curs_write(&prep, smc_curs_read(&conn->tx_curs_prep, conn), conn);
	311	/* cf. wmem_alloc - (snd_max - snd_una) */
69cb7dc0	312	to_send = smc_curs_diff(conn->sndbuf_desc->len, &sent, &prep);
e6727f39 UB	313	if (to_send <= 0)
	314	return 0;
	315
	316	/* destination: RMBE */
	317	/* cf. snd_wnd */
	318	rmbespace = atomic_read(&conn->peer_rmbe_space);
	319	if (rmbespace <= 0)
	320	return 0;
	321	smc_curs_write(&prod,
	322	smc_curs_read(&conn->local_tx_ctrl.prod, conn),
	323	conn);
	324	smc_curs_write(&cons,
	325	smc_curs_read(&conn->local_rx_ctrl.cons, conn),
	326	conn);
	327
	328	/* if usable snd_wnd closes ask peer to advertise once it opens again */
	329	conn->local_tx_ctrl.prod_flags.write_blocked = (to_send >= rmbespace);
	330	/* cf. usable snd_wnd */
	331	len = min(to_send, rmbespace);
	332
	333	/* initialize variables for first iteration of subsequent nested loop */
	334	link = &lgr->lnk[SMC_SINGLE_LINK];
	335	dst_off = prod.count;
	336	if (prod.wrap == cons.wrap) {
	337	/* the filled destination area is unwrapped,
	338	* hence the available free destination space is wrapped
	339	* and we need 2 destination chunks of sum len; start with 1st
	340	* which is limited by what's available in sndbuf
	341	*/
	342	dst_len = min_t(size_t,
	343	conn->peer_rmbe_size - prod.count, len);
	344	} else {
	345	/* the filled destination area is wrapped,
	346	* hence the available free destination space is unwrapped
	347	* and we need a single destination chunk of entire len
	348	*/
	349	dst_len = len;
	350	}
	351	dst_len_sum = dst_len;
	352	src_off = sent.count;
	353	/* dst_len determines the maximum src_len */
69cb7dc0	354	if (sent.count + dst_len <= conn->sndbuf_desc->len) {
e6727f39 UB	355	/* unwrapped src case: single chunk of entire dst_len */
	356	src_len = dst_len;
	357	} else {
	358	/* wrapped src case: 2 chunks of sum dst_len; start with 1st: */
69cb7dc0	359	src_len = conn->sndbuf_desc->len - sent.count;
e6727f39 UB	360	}
e6727f39 UB	361	src_len_sum = src_len;
9d8fb617	362	dma_addr = sg_dma_address(conn->sndbuf_desc->sgt[SMC_SINGLE_LINK].sgl);
e6727f39 UB	363	for (dstchunk = 0; dstchunk < 2; dstchunk++) {
	364	num_sges = 0;
	365	for (srcchunk = 0; srcchunk < 2; srcchunk++) {
9d8fb617	366	sges[srcchunk].addr = dma_addr + src_off;
e6727f39 UB	367	sges[srcchunk].length = src_len;
	368	sges[srcchunk].lkey = link->roce_pd->local_dma_lkey;
	369	num_sges++;
	370	src_off += src_len;
69cb7dc0 HW	371	if (src_off >= conn->sndbuf_desc->len)
69cb7dc0 HW	372	src_off -= conn->sndbuf_desc->len;
e6727f39 UB	373	/* modulo in send ring */
	374	if (src_len_sum == dst_len)
	375	break; /* either on 1st or 2nd iteration */
	376	/* prepare next (== 2nd) iteration */
	377	src_len = dst_len - src_len; /* remainder */
	378	src_len_sum += src_len;
	379	}
	380	rc = smc_tx_rdma_write(conn, dst_off, num_sges, sges);
	381	if (rc)
	382	return rc;
	383	if (dst_len_sum == len)
	384	break; /* either on 1st or 2nd iteration */
	385	/* prepare next (== 2nd) iteration */
	386	dst_off = 0; /* modulo offset in RMBE ring buffer */
	387	dst_len = len - dst_len; /* remainder */
	388	dst_len_sum += dst_len;
	389	src_len = min_t(int,
69cb7dc0	390	dst_len, conn->sndbuf_desc->len - sent.count);
e6727f39 UB	391	src_len_sum = src_len;
	392	}
	393
	394	smc_tx_advance_cursors(conn, &prod, &sent, len);
	395	/* update connection's cursors with advanced local cursors */
	396	smc_curs_write(&conn->local_tx_ctrl.prod,
	397	smc_curs_read(&prod, conn),
	398	conn);
	399	/* dst: peer RMBE */
	400	smc_curs_write(&conn->tx_curs_sent,
	401	smc_curs_read(&sent, conn),
	402	conn);
	403	/* src: local sndbuf */
	404
	405	return 0;
	406	}
	407
	408	/* Wakeup sndbuf consumers from any context (IRQ or process)
	409	* since there is more data to transmit; usable snd_wnd as max transmit
	410	*/
	411	int smc_tx_sndbuf_nonempty(struct smc_connection *conn)
	412	{
	413	struct smc_cdc_tx_pend *pend;
	414	struct smc_wr_buf *wr_buf;
	415	int rc;
	416
	417	spin_lock_bh(&conn->send_lock);
51957bc5	418	rc = smc_cdc_get_free_slot(conn, &wr_buf, &pend);
e6727f39 UB	419	if (rc < 0) {
e6727f39 UB	420	if (rc == -EBUSY) {
b38d7324 UB	421	struct smc_sock *smc =
	422	container_of(conn, struct smc_sock, conn);
	423
	424	if (smc->sk.sk_err == ECONNABORTED) {
	425	rc = sock_error(&smc->sk);
	426	goto out_unlock;
	427	}
e6727f39	428	rc = 0;
1a0a04c7	429	if (conn->alert_token_local) /* connection healthy */
01d2f7e2 UB	430	mod_delayed_work(system_wq, &conn->tx_work,
01d2f7e2 UB	431	SMC_TX_WORK_DELAY);
e6727f39 UB	432	}
	433	goto out_unlock;
	434	}
	435
	436	rc = smc_tx_rdma_writes(conn);
	437	if (rc) {
	438	smc_wr_tx_put_slot(&conn->lgr->lnk[SMC_SINGLE_LINK],
	439	(struct smc_wr_tx_pend_priv *)pend);
	440	goto out_unlock;
	441	}
	442
	443	rc = smc_cdc_msg_send(conn, wr_buf, pend);
	444
	445	out_unlock:
	446	spin_unlock_bh(&conn->send_lock);
	447	return rc;
	448	}
	449
	450	/* Wakeup sndbuf consumers from process context
	451	* since there is more data to transmit
	452	*/
be7f3e59	453	void smc_tx_work(struct work_struct *work)
e6727f39	454	{
18e537cd	455	struct smc_connection *conn = container_of(to_delayed_work(work),
e6727f39 UB	456	struct smc_connection,
	457	tx_work);
	458	struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
90cacb2e	459	int rc;
e6727f39 UB	460
e6727f39 UB	461	lock_sock(&smc->sk);
1a0a04c7 UB	462	if (smc->sk.sk_err \|\|
	463	!conn->alert_token_local \|\|
	464	conn->local_rx_ctrl.conn_state_flags.peer_conn_abort)
	465	goto out;
	466
90cacb2e UB	467	rc = smc_tx_sndbuf_nonempty(conn);
	468	if (!rc && conn->local_rx_ctrl.prod_flags.write_blocked &&
	469	!atomic_read(&conn->bytes_to_rcv))
	470	conn->local_rx_ctrl.prod_flags.write_blocked = 0;
1a0a04c7 UB	471
1a0a04c7 UB	472	out:
e6727f39 UB	473	release_sock(&smc->sk);
	474	}
	475
952310cc UB	476	void smc_tx_consumer_update(struct smc_connection *conn)
	477	{
	478	union smc_host_cursor cfed, cons;
6b5771aa	479	int to_confirm;
952310cc UB	480
	481	smc_curs_write(&cons,
	482	smc_curs_read(&conn->local_tx_ctrl.cons, conn),
	483	conn);
	484	smc_curs_write(&cfed,
	485	smc_curs_read(&conn->rx_curs_confirmed, conn),
	486	conn);
69cb7dc0	487	to_confirm = smc_curs_diff(conn->rmb_desc->len, &cfed, &cons);
952310cc UB	488
	489	if (conn->local_rx_ctrl.prod_flags.cons_curs_upd_req \|\|
	490	((to_confirm > conn->rmbe_update_limit) &&
69cb7dc0	491	((to_confirm > (conn->rmb_desc->len / 2)) \|\|
952310cc	492	conn->local_rx_ctrl.prod_flags.write_blocked))) {
1a0a04c7 UB	493	if ((smc_cdc_get_slot_and_msg_send(conn) < 0) &&
1a0a04c7 UB	494	conn->alert_token_local) { /* connection healthy */
18e537cd UB	495	schedule_delayed_work(&conn->tx_work,
18e537cd UB	496	SMC_TX_WORK_DELAY);
952310cc UB	497	return;
	498	}
	499	smc_curs_write(&conn->rx_curs_confirmed,
	500	smc_curs_read(&conn->local_tx_ctrl.cons, conn),
	501	conn);
	502	conn->local_rx_ctrl.prod_flags.cons_curs_upd_req = 0;
	503	}
	504	if (conn->local_rx_ctrl.prod_flags.write_blocked &&
	505	!atomic_read(&conn->bytes_to_rcv))
	506	conn->local_rx_ctrl.prod_flags.write_blocked = 0;
	507	}
	508
e6727f39 UB	509	/*************************** send initialize *****************************/
	510
	511	/* Initialize send properties on connection establishment. NB: not __init! */
	512	void smc_tx_init(struct smc_sock *smc)
	513	{
	514	smc->sk.sk_write_space = smc_tx_write_space;
e6727f39	515	}