iscsi_tcp.c

	/*
	 * Use rd_desc to pass 'conn' to iscsi_tcp_recv.
	 * We set count to 1 because we want the network layer to
	 * hand us all the skbs that are available. iscsi_tcp_recv
	 * handled pdus that cross buffers or pdus that still need data.
	 */
	rd_desc.arg.data = conn;
	rd_desc.count = 1;
	tcp_read_sock(sk, &rd_desc, iscsi_tcp_recv);

	read_unlock(&sk->sk_callback_lock);

	/* If we had to (atomically) map a highmem page,
	 * unmap it now. */
	iscsi_tcp_segment_unmap(&tcp_conn->in.segment);
}

static void
iscsi_tcp_state_change(struct sock *sk)
{
	struct iscsi_tcp_conn *tcp_conn;
	struct iscsi_conn *conn;
	struct iscsi_session *session;
	void (*old_state_change)(struct sock *);

	read_lock(&sk->sk_callback_lock);

	conn = (struct iscsi_conn*)sk->sk_user_data;
	session = conn->session;

	if ((sk->sk_state == TCP_CLOSE_WAIT ||
	     sk->sk_state == TCP_CLOSE) &&
	    !atomic_read(&sk->sk_rmem_alloc)) {
		debug_tcp("iscsi_tcp_state_change: TCP_CLOSE|TCP_CLOSE_WAIT\n");
		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
	}

	tcp_conn = conn->dd_data;
	old_state_change = tcp_conn->old_state_change;

	read_unlock(&sk->sk_callback_lock);

	old_state_change(sk);
}

/**
 * iscsi_write_space - Called when more output buffer space is available
 * @sk: socket space is available for
 **/
static void
iscsi_write_space(struct sock *sk)
{
	struct iscsi_conn *conn = (struct iscsi_conn*)sk->sk_user_data;
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;

	tcp_conn->old_write_space(sk);
	debug_tcp("iscsi_write_space: cid %d\n", conn->id);
	scsi_queue_work(conn->session->host, &conn->xmitwork);
}

static void
iscsi_conn_set_callbacks(struct iscsi_conn *conn)
{
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct sock *sk = tcp_conn->sock->sk;

	/* assign new callbacks */
	write_lock_bh(&sk->sk_callback_lock);
	sk->sk_user_data = conn;
	tcp_conn->old_data_ready = sk->sk_data_ready;
	tcp_conn->old_state_change = sk->sk_state_change;
	tcp_conn->old_write_space = sk->sk_write_space;
	sk->sk_data_ready = iscsi_tcp_data_ready;
	sk->sk_state_change = iscsi_tcp_state_change;
	sk->sk_write_space = iscsi_write_space;
	write_unlock_bh(&sk->sk_callback_lock);
}

static void
iscsi_conn_restore_callbacks(struct iscsi_tcp_conn *tcp_conn)
{
	struct sock *sk = tcp_conn->sock->sk;

	/* restore socket callbacks, see also: iscsi_conn_set_callbacks() */
	write_lock_bh(&sk->sk_callback_lock);
	sk->sk_user_data    = NULL;
	sk->sk_data_ready   = tcp_conn->old_data_ready;
	sk->sk_state_change = tcp_conn->old_state_change;
	sk->sk_write_space  = tcp_conn->old_write_space;
	sk->sk_no_check	 = 0;
	write_unlock_bh(&sk->sk_callback_lock);
}

/**
 * iscsi_xmit - TCP transmit
 **/
static int
iscsi_xmit(struct iscsi_conn *conn)
{
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct iscsi_segment *segment = &tcp_conn->out.segment;
	unsigned int consumed = 0;
	int rc = 0;

	while (1) {
		rc = iscsi_tcp_xmit_segment(tcp_conn, segment);
		if (rc < 0)
			goto error;
		if (rc == 0)
			break;

		consumed += rc;

		if (segment->total_copied >= segment->total_size) {
			if (segment->done != NULL) {
				rc = segment->done(tcp_conn, segment);
				if (rc < 0)
					goto error;
			}
		}
	}

	debug_tcp("xmit %d bytes\n", consumed);

	conn->txdata_octets += consumed;
	return consumed;

error:
	/* Transmit error. We could initiate error recovery
	 * here. */
	debug_tcp("Error sending PDU, errno=%d\n", rc);
	iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
	return rc;
}

/**
 * iscsi_tcp_xmit_qlen - return the number of bytes queued for xmit
 */
static inline int
iscsi_tcp_xmit_qlen(struct iscsi_conn *conn)
{
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct iscsi_segment *segment = &tcp_conn->out.segment;

	return segment->total_copied - segment->total_size;
}

static inline int
iscsi_tcp_flush(struct iscsi_conn *conn)
{
	int rc;

	while (iscsi_tcp_xmit_qlen(conn)) {
		rc = iscsi_xmit(conn);
		if (rc == 0)
			return -EAGAIN;
		if (rc < 0)
			return rc;
	}

	return 0;
}

/*
 * This is called when we're done sending the header.
 * Simply copy the data_segment to the send segment, and return.
 */
static int
iscsi_tcp_send_hdr_done(struct iscsi_tcp_conn *tcp_conn,
			struct iscsi_segment *segment)
{
	tcp_conn->out.segment = tcp_conn->out.data_segment;
	debug_tcp("Header done. Next segment size %u total_size %u\n",
		  tcp_conn->out.segment.size, tcp_conn->out.segment.total_size);
	return 0;
}

static void
iscsi_tcp_send_hdr_prep(struct iscsi_conn *conn, void *hdr, size_t hdrlen)
{
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;

	debug_tcp("%s(%p%s)\n", __FUNCTION__, tcp_conn,
			conn->hdrdgst_en? ", digest enabled" : "");

	/* Clear the data segment - needs to be filled in by the
	 * caller using iscsi_tcp_send_data_prep() */
	memset(&tcp_conn->out.data_segment, 0, sizeof(struct iscsi_segment));

	/* If header digest is enabled, compute the CRC and
	 * place the digest into the same buffer. We make
	 * sure that both iscsi_tcp_ctask and mtask have
	 * sufficient room.
	 */
	if (conn->hdrdgst_en) {
		iscsi_tcp_dgst_header(&tcp_conn->tx_hash, hdr, hdrlen,
				      hdr + hdrlen);
		hdrlen += ISCSI_DIGEST_SIZE;
	}

	/* Remember header pointer for later, when we need
	 * to decide whether there's a payload to go along
	 * with the header. */
	tcp_conn->out.hdr = hdr;

	iscsi_segment_init_linear(&tcp_conn->out.segment, hdr, hdrlen,
				iscsi_tcp_send_hdr_done, NULL);
}

/*
 * Prepare the send buffer for the payload data.
 * Padding and checksumming will all be taken care
 * of by the iscsi_segment routines.
 */
static int
iscsi_tcp_send_data_prep(struct iscsi_conn *conn, struct scatterlist *sg,
			 unsigned int count, unsigned int offset,
			 unsigned int len)
{
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct hash_desc *tx_hash = NULL;
	unsigned int hdr_spec_len;

	debug_tcp("%s(%p, offset=%d, datalen=%d%s)\n", __FUNCTION__,
			tcp_conn, offset, len,
			conn->datadgst_en? ", digest enabled" : "");

	/* Make sure the datalen matches what the caller
	   said he would send. */
	hdr_spec_len = ntoh24(tcp_conn->out.hdr->dlength);
	WARN_ON(iscsi_padded(len) != iscsi_padded(hdr_spec_len));

	if (conn->datadgst_en)
		tx_hash = &tcp_conn->tx_hash;

	return iscsi_segment_seek_sg(&tcp_conn->out.data_segment,
				   sg, count, offset, len,
				   NULL, tx_hash);
}

static void
iscsi_tcp_send_linear_data_prepare(struct iscsi_conn *conn, void *data,
				   size_t len)
{
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct hash_desc *tx_hash = NULL;
	unsigned int hdr_spec_len;

	debug_tcp("%s(%p, datalen=%d%s)\n", __FUNCTION__, tcp_conn, len,
		  conn->datadgst_en? ", digest enabled" : "");

	/* Make sure the datalen matches what the caller
	   said he would send. */
	hdr_spec_len = ntoh24(tcp_conn->out.hdr->dlength);
	WARN_ON(iscsi_padded(len) != iscsi_padded(hdr_spec_len));

	if (conn->datadgst_en)
		tx_hash = &tcp_conn->tx_hash;

	iscsi_segment_init_linear(&tcp_conn->out.data_segment,
				data, len, NULL, tx_hash);
}

/**
 * iscsi_solicit_data_cont - initialize next Data-Out
 * @conn: iscsi connection
 * @ctask: scsi command task
 * @r2t: R2T info
 * @left: bytes left to transfer
 *
 * Notes:
 *	Initialize next Data-Out within this R2T sequence and continue
 *	to process next Scatter-Gather element(if any) of this SCSI command.
 *
 *	Called under connection lock.
 **/
static int
iscsi_solicit_data_cont(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
			struct iscsi_r2t_info *r2t)
{
	struct iscsi_data *hdr;
	int new_offset, left;

	BUG_ON(r2t->data_length - r2t->sent < 0);
	left = r2t->data_length - r2t->sent;
	if (left == 0)
		return 0;

	hdr = &r2t->dtask.hdr;
	memset(hdr, 0, sizeof(struct iscsi_data));
	hdr->ttt = r2t->ttt;
	hdr->datasn = cpu_to_be32(r2t->solicit_datasn);
	r2t->solicit_datasn++;
	hdr->opcode = ISCSI_OP_SCSI_DATA_OUT;
	memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun));
	hdr->itt = ctask->hdr->itt;
	hdr->exp_statsn = r2t->exp_statsn;
	new_offset = r2t->data_offset + r2t->sent;
	hdr->offset = cpu_to_be32(new_offset);
	if (left > conn->max_xmit_dlength) {
		hton24(hdr->dlength, conn->max_xmit_dlength);
		r2t->data_count = conn->max_xmit_dlength;
	} else {
		hton24(hdr->dlength, left);
		r2t->data_count = left;
		hdr->flags = ISCSI_FLAG_CMD_FINAL;
	}

	conn->dataout_pdus_cnt++;
	return 1;
}

/**
 * iscsi_tcp_ctask - Initialize iSCSI SCSI_READ or SCSI_WRITE commands
 * @conn: iscsi connection
 * @ctask: scsi command task
 * @sc: scsi command
 **/
static int
iscsi_tcp_ctask_init(struct iscsi_cmd_task *ctask)
{
	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
	struct iscsi_conn *conn = ctask->conn;
	struct scsi_cmnd *sc = ctask->sc;
	int err;

	BUG_ON(__kfifo_len(tcp_ctask->r2tqueue));
	tcp_ctask->sent = 0;
	tcp_ctask->exp_datasn = 0;

	/* Prepare PDU, optionally w/ immediate data */
	debug_scsi("ctask deq [cid %d itt 0x%x imm %d unsol %d]\n",
		    conn->id, ctask->itt, ctask->imm_count,
		    ctask->unsol_count);
	iscsi_tcp_send_hdr_prep(conn, ctask->hdr, ctask->hdr_len);

	if (!ctask->imm_count)
		return 0;

	/* If we have immediate data, attach a payload */
	err = iscsi_tcp_send_data_prep(conn, scsi_out(sc)->table.sgl,
				       scsi_out(sc)->table.nents,
				       0, ctask->imm_count);
	if (err)
		return err;
	tcp_ctask->sent += ctask->imm_count;
	ctask->imm_count = 0;
	return 0;
}

/**
 * iscsi_tcp_mtask_xmit - xmit management(immediate) task
 * @conn: iscsi connection
 * @mtask: task management task
 *
 * Notes:
 *	The function can return -EAGAIN in which case caller must
 *	call it again later, or recover. '0' return code means successful
 *	xmit.
 **/
static int
iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
{
	int rc;

	/* Flush any pending data first. */
	rc = iscsi_tcp_flush(conn);
	if (rc < 0)
		return rc;

	if (mtask->hdr->itt == RESERVED_ITT) {
		struct iscsi_session *session = conn->session;

		spin_lock_bh(&session->lock);
		iscsi_free_mgmt_task(conn, mtask);
		spin_unlock_bh(&session->lock);
	}

	return 0;
}

/*
 * iscsi_tcp_ctask_xmit - xmit normal PDU task
 * @conn: iscsi connection
 * @ctask: iscsi command task
 *
 * We're expected to return 0 when everything was transmitted succesfully,
 * -EAGAIN if there's still data in the queue, or != 0 for any other kind
 * of error.
 */
static int
iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
{
	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
	struct scsi_cmnd *sc = ctask->sc;
	struct scsi_data_buffer *sdb = scsi_out(sc);
	int rc = 0;

flush:
	/* Flush any pending data first. */
	rc = iscsi_tcp_flush(conn);
	if (rc < 0)
		return rc;

	/* Are we done already? */
	if (sc->sc_data_direction != DMA_TO_DEVICE)
		return 0;

	if (ctask->unsol_count != 0) {
		struct iscsi_data *hdr = &tcp_ctask->unsol_dtask.hdr;

		/* Prepare a header for the unsolicited PDU.
		 * The amount of data we want to send will be
		 * in ctask->data_count.
		 * FIXME: return the data count instead.
		 */
		iscsi_prep_unsolicit_data_pdu(ctask, hdr);

		debug_tcp("unsol dout [itt 0x%x doff %d dlen %d]\n",
				ctask->itt, tcp_ctask->sent, ctask->data_count);

		iscsi_tcp_send_hdr_prep(conn, hdr, sizeof(*hdr));
		rc = iscsi_tcp_send_data_prep(conn, sdb->table.sgl,
					      sdb->table.nents, tcp_ctask->sent,
					      ctask->data_count);
		if (rc)
			goto fail;
		tcp_ctask->sent += ctask->data_count;
		ctask->unsol_count -= ctask->data_count;
		goto flush;
	} else {
		struct iscsi_session *session = conn->session;
		struct iscsi_r2t_info *r2t;

		/* All unsolicited PDUs sent. Check for solicited PDUs.
		 */
		spin_lock_bh(&session->lock);
		r2t = tcp_ctask->r2t;
		if (r2t != NULL) {
			/* Continue with this R2T? */
			if (!iscsi_solicit_data_cont(conn, ctask, r2t)) {
				debug_scsi("  done with r2t %p\n", r2t);

				__kfifo_put(tcp_ctask->r2tpool.queue,
					    (void*)&r2t, sizeof(void*));
				tcp_ctask->r2t = r2t = NULL;
			}
		}

		if (r2t == NULL) {
			__kfifo_get(tcp_ctask->r2tqueue, (void*)&tcp_ctask->r2t,
				    sizeof(void*));
			r2t = tcp_ctask->r2t;
		}
		spin_unlock_bh(&session->lock);

		/* Waiting for more R2Ts to arrive. */
		if (r2t == NULL) {
			debug_tcp("no R2Ts yet\n");
			return 0;
		}

		debug_scsi("sol dout %p [dsn %d itt 0x%x doff %d dlen %d]\n",
			r2t, r2t->solicit_datasn - 1, ctask->itt,
			r2t->data_offset + r2t->sent, r2t->data_count);

		iscsi_tcp_send_hdr_prep(conn, &r2t->dtask.hdr,
					sizeof(struct iscsi_hdr));

		rc = iscsi_tcp_send_data_prep(conn, sdb->table.sgl,
					      sdb->table.nents,
					      r2t->data_offset + r2t->sent,
					      r2t->data_count);
		if (rc)
			goto fail;
		tcp_ctask->sent += r2t->data_count;
		r2t->sent += r2t->data_count;
		goto flush;
	}
	return 0;
fail:
	iscsi_conn_failure(conn, rc);
	return -EIO;
}

static struct iscsi_cls_conn *
iscsi_tcp_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
{
	struct iscsi_conn *conn;
	struct iscsi_cls_conn *cls_conn;
	struct iscsi_tcp_conn *tcp_conn;

	cls_conn = iscsi_conn_setup(cls_session, conn_idx);
	if (!cls_conn)
		return NULL;
	conn = cls_conn->dd_data;
	/*
	 * due to strange issues with iser these are not set
	 * in iscsi_conn_setup
	 */
	conn->max_recv_dlength = ISCSI_DEF_MAX_RECV_SEG_LEN;

	tcp_conn = kzalloc(sizeof(*tcp_conn), GFP_KERNEL);
	if (!tcp_conn)
		goto tcp_conn_alloc_fail;

	conn->dd_data = tcp_conn;
	tcp_conn->iscsi_conn = conn;

	tcp_conn->tx_hash.tfm = crypto_alloc_hash("crc32c", 0,
						  CRYPTO_ALG_ASYNC);
	tcp_conn->tx_hash.flags = 0;
	if (IS_ERR(tcp_conn->tx_hash.tfm))
		goto free_tcp_conn;

	tcp_conn->rx_hash.tfm = crypto_alloc_hash("crc32c", 0,
						  CRYPTO_ALG_ASYNC);
	tcp_conn->rx_hash.flags = 0;
	if (IS_ERR(tcp_conn->rx_hash.tfm))
		goto free_tx_tfm;

	return cls_conn;

free_tx_tfm:
	crypto_free_hash(tcp_conn->tx_hash.tfm);
free_tcp_conn:
	iscsi_conn_printk(KERN_ERR, conn,
			  "Could not create connection due to crc32c "
			  "loading error. Make sure the crc32c "
			  "module is built as a module or into the "
			  "kernel\n");
	kfree(tcp_conn);
tcp_conn_alloc_fail:
	iscsi_conn_teardown(cls_conn);
	return NULL;
}

static void
iscsi_tcp_release_conn(struct iscsi_conn *conn)
{
	struct iscsi_session *session = conn->session;
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct socket *sock = tcp_conn->sock;

	if (!sock)
		return;

	sock_hold(sock->sk);
	iscsi_conn_restore_callbacks(tcp_conn);
	sock_put(sock->sk);

	spin_lock_bh(&session->lock);
	tcp_conn->sock = NULL;
	conn->recv_lock = NULL;
	spin_unlock_bh(&session->lock);
	sockfd_put(sock);
}

static void
iscsi_tcp_conn_destroy(struct iscsi_cls_conn *cls_conn)
{
	struct iscsi_conn *conn = cls_conn->dd_data;
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;

	iscsi_tcp_release_conn(conn);
	iscsi_conn_teardown(cls_conn);

	if (tcp_conn->tx_hash.tfm)
		crypto_free_hash(tcp_conn->tx_hash.tfm);
	if (tcp_conn->rx_hash.tfm)
		crypto_free_hash(tcp_conn->rx_hash.tfm);

	kfree(tcp_conn);
}

static void
iscsi_tcp_conn_stop(struct iscsi_cls_conn *cls_conn, int flag)
{
	struct iscsi_conn *conn = cls_conn->dd_data;

	iscsi_conn_stop(cls_conn, flag);
	iscsi_tcp_release_conn(conn);
}

static int iscsi_tcp_get_addr(struct iscsi_conn *conn, struct socket *sock,
			      char *buf, int *port,
			      int (*getname)(struct socket *, struct sockaddr *,
					int *addrlen))
{
	struct sockaddr_storage *addr;
	struct sockaddr_in6 *sin6;
	struct sockaddr_in *sin;
	int rc = 0, len;

	addr = kmalloc(sizeof(*addr), GFP_KERNEL);
	if (!addr)
		return -ENOMEM;

	if (getname(sock, (struct sockaddr *) addr, &len)) {
		rc = -ENODEV;
		goto free_addr;
	}

	switch (addr->ss_family) {
	case AF_INET:
		sin = (struct sockaddr_in *)addr;
		spin_lock_bh(&conn->session->lock);
		sprintf(buf, NIPQUAD_FMT, NIPQUAD(sin->sin_addr.s_addr));
		*port = be16_to_cpu(sin->sin_port);
		spin_unlock_bh(&conn->session->lock);
		break;
	case AF_INET6:
		sin6 = (struct sockaddr_in6 *)addr;
		spin_lock_bh(&conn->session->lock);
		sprintf(buf, NIP6_FMT, NIP6(sin6->sin6_addr));
		*port = be16_to_cpu(sin6->sin6_port);
		spin_unlock_bh(&conn->session->lock);
		break;
	}
free_addr:
	kfree(addr);
	return rc;
}

static int
iscsi_tcp_conn_bind(struct iscsi_cls_session *cls_session,
		    struct iscsi_cls_conn *cls_conn, uint64_t transport_eph,
		    int is_leading)
{
	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);
	struct iscsi_host *ihost = shost_priv(shost);
	struct iscsi_conn *conn = cls_conn->dd_data;
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	struct sock *sk;
	struct socket *sock;
	int err;

	/* lookup for existing socket */
	sock = sockfd_lookup((int)transport_eph, &err);
	if (!sock) {
		iscsi_conn_printk(KERN_ERR, conn,
				  "sockfd_lookup failed %d\n", err);
		return -EEXIST;
	}
	/*
	 * copy these values now because if we drop the session
	 * userspace may still want to query the values since we will
	 * be using them for the reconnect
	 */
	err = iscsi_tcp_get_addr(conn, sock, conn->portal_address,
				 &conn->portal_port, kernel_getpeername);
	if (err)
		goto free_socket;

	err = iscsi_tcp_get_addr(conn, sock, ihost->local_address,
				&ihost->local_port, kernel_getsockname);
	if (err)
		goto free_socket;

	err = iscsi_conn_bind(cls_session, cls_conn, is_leading);
	if (err)
		goto free_socket;

	/* bind iSCSI connection and socket */
	tcp_conn->sock = sock;

	/* setup Socket parameters */
	sk = sock->sk;
	sk->sk_reuse = 1;
	sk->sk_sndtimeo = 15 * HZ; /* FIXME: make it configurable */
	sk->sk_allocation = GFP_ATOMIC;

	/* FIXME: disable Nagle's algorithm */

	/*
	 * Intercept TCP callbacks for sendfile like receive
	 * processing.
	 */
	conn->recv_lock = &sk->sk_callback_lock;
	iscsi_conn_set_callbacks(conn);
	tcp_conn->sendpage = tcp_conn->sock->ops->sendpage;
	/*
	 * set receive state machine into initial state
	 */
	iscsi_tcp_hdr_recv_prep(tcp_conn);
	return 0;

free_socket:
	sockfd_put(sock);
	return err;
}

/* called with host lock */
static void
iscsi_tcp_mtask_init(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
{
	debug_scsi("mtask deq [cid %d itt 0x%x]\n", conn->id, mtask->itt);

	/* Prepare PDU, optionally w/ immediate data */
	iscsi_tcp_send_hdr_prep(conn, mtask->hdr, sizeof(*mtask->hdr));

	/* If we have immediate data, attach a payload */
	if (mtask->data_count)
		iscsi_tcp_send_linear_data_prepare(conn, mtask->data,
						   mtask->data_count);
}

static int
iscsi_r2tpool_alloc(struct iscsi_session *session)
{
	int i;
	int cmd_i;

	/*
	 * initialize per-task: R2T pool and xmit queue
	 */
	for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) {
	        struct iscsi_cmd_task *ctask = session->cmds[cmd_i];
		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;

		/*
		 * pre-allocated x4 as much r2ts to handle race when
		 * target acks DataOut faster than we data_xmit() queues
		 * could replenish r2tqueue.
		 */

		/* R2T pool */
		if (iscsi_pool_init(&tcp_ctask->r2tpool, session->max_r2t * 4, NULL,
				    sizeof(struct iscsi_r2t_info))) {
			goto r2t_alloc_fail;
		}

		/* R2T xmit queue */
		tcp_ctask->r2tqueue = kfifo_alloc(
		      session->max_r2t * 4 * sizeof(void*), GFP_KERNEL, NULL);
		if (tcp_ctask->r2tqueue == ERR_PTR(-ENOMEM)) {
			iscsi_pool_free(&tcp_ctask->r2tpool);
			goto r2t_alloc_fail;
		}
	}

	return 0;

r2t_alloc_fail:
	for (i = 0; i < cmd_i; i++) {
		struct iscsi_cmd_task *ctask = session->cmds[i];
		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;

		kfifo_free(tcp_ctask->r2tqueue);
		iscsi_pool_free(&tcp_ctask->r2tpool);
	}
	return -ENOMEM;
}

static void
iscsi_r2tpool_free(struct iscsi_session *session)
{
	int i;

	for (i = 0; i < session->cmds_max; i++) {
		struct iscsi_cmd_task *ctask = session->cmds[i];
		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;

		kfifo_free(tcp_ctask->r2tqueue);
		iscsi_pool_free(&tcp_ctask->r2tpool);
	}
}

static int
iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param,
		     char *buf, int buflen)
{
	struct iscsi_conn *conn = cls_conn->dd_data;
	struct iscsi_session *session = conn->session;
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
	int value;

	switch(param) {
	case ISCSI_PARAM_HDRDGST_EN:
		iscsi_set_param(cls_conn, param, buf, buflen);
		break;
	case ISCSI_PARAM_DATADGST_EN:
		iscsi_set_param(cls_conn, param, buf, buflen);
		tcp_conn->sendpage = conn->datadgst_en ?
			sock_no_sendpage : tcp_conn->sock->ops->sendpage;
		break;
	case ISCSI_PARAM_MAX_R2T:
		sscanf(buf, "%d", &value);
		if (value <= 0 || !is_power_of_2(value))
			return -EINVAL;
		if (session->max_r2t == value)
			break;
		iscsi_r2tpool_free(session);
		iscsi_set_param(cls_conn, param, buf, buflen);
		if (iscsi_r2tpool_alloc(session))
			return -ENOMEM;
		break;
	default:
		return iscsi_set_param(cls_conn, param, buf, buflen);
	}

	return 0;
}

static int
iscsi_tcp_conn_get_param(struct iscsi_cls_conn *cls_conn,
			 enum iscsi_param param, char *buf)
{
	struct iscsi_conn *conn = cls_conn->dd_data;
	int len;

	switch(param) {
	case ISCSI_PARAM_CONN_PORT:
		spin_lock_bh(&conn->session->lock);
		len = sprintf(buf, "%hu\n", conn->portal_port);
		spin_unlock_bh(&conn->session->lock);
		break;
	case ISCSI_PARAM_CONN_ADDRESS:
		spin_lock_bh(&conn->session->lock);
		len = sprintf(buf, "%s\n", conn->portal_address);
		spin_unlock_bh(&conn->session->lock);
		break;
	default:
		return iscsi_conn_get_param(cls_conn, param, buf);
	}

	return len;
}

static void
iscsi_conn_get_stats(struct iscsi_cls_conn *cls_conn, struct iscsi_stats *stats)
{
	struct iscsi_conn *conn = cls_conn->dd_data;
	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;

	stats->txdata_octets = conn->txdata_octets;
	stats->rxdata_octets = conn->rxdata_octets;
	stats->scsicmd_pdus = conn->scsicmd_pdus_cnt;
	stats->dataout_pdus = conn->dataout_pdus_cnt;
	stats->scsirsp_pdus = conn->scsirsp_pdus_cnt;
	stats->datain_pdus = conn->datain_pdus_cnt;
	stats->r2t_pdus = conn->r2t_pdus_cnt;
	stats->tmfcmd_pdus = conn->tmfcmd_pdus_cnt;
	stats->tmfrsp_pdus = conn->tmfrsp_pdus_cnt;
	stats->custom_length = 3;
	strcpy(stats->custom[0].desc, "tx_sendpage_failures");
	stats->custom[0].value = tcp_conn->sendpage_failures_cnt;
	strcpy(stats->custom[1].desc, "rx_discontiguous_hdr");
	stats->custom[1].value = tcp_conn->discontiguous_hdr_cnt;
	strcpy(stats->custom[2].desc, "eh_abort_cnt");
	stats->custom[2].value = conn->eh_abort_cnt;
}

static struct iscsi_cls_session *
iscsi_tcp_session_create(struct Scsi_Host *shost, uint16_t cmds_max,
			 uint16_t qdepth, uint32_t initial_cmdsn,
			 uint32_t *hostno)
{
	struct iscsi_cls_session *cls_session;
	struct iscsi_session *session;
	int cmd_i;

	if (shost) {
		printk(KERN_ERR "iscsi_tcp: invalid shost %d.\n",
		       shost->host_no);
		return NULL;
	}

	shost = iscsi_host_alloc(&iscsi_sht, 0, qdepth);
	if (!shost)
		return NULL;
	shost->transportt = iscsi_tcp_scsi_transport;
	shost->max_lun = iscsi_max_lun;
	shost->max_id = 0;
	shost->max_channel = 0;
	shost->max_cmd_len = 16;
	shost->can_queue = cmds_max;

	if (iscsi_host_add(shost, NULL))
		goto free_host;
	*hostno = shost->host_no;

	cls_session = iscsi_session_setup(&iscsi_tcp_transport, shost, cmds_max,
					  sizeof(struct iscsi_tcp_cmd_task),
					  sizeof(struct iscsi_tcp_mgmt_task),
					  initial_cmdsn);
	if (!cls_session)
		goto remove_host;
	session = cls_session->dd_data;

	shost->can_queue = session->cmds_max;
	for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) {
		struct iscsi_cmd_task *ctask = session->cmds[cmd_i];
		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;

		ctask->hdr = &tcp_ctask->hdr.cmd_hdr;
		ctask->hdr_max = sizeof(tcp_ctask->hdr) - ISCSI_DIGEST_SIZE;
	}

	for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) {
		struct iscsi_mgmt_task *mtask = session->mgmt_cmds[cmd_i];
		struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data;

		mtask->hdr = (struct iscsi_hdr *) &tcp_mtask->hdr;
	}

	if (iscsi_r2tpool_alloc(session))
		goto remove_session;
	return cls_session;

remove_session:
	iscsi_session_teardown(cls_session);
remove_host:
	iscsi_host_remove(shost);
free_host:
	iscsi_host_free(shost);
	return NULL;
}

static void iscsi_tcp_session_destroy(struct iscsi_cls_session *cls_session)
{
	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);

	iscsi_r2tpool_free(cls_session->dd_data);

	iscsi_host_remove(shost);
	iscsi_host_free(shost);
}

static int iscsi_tcp_slave_configure(struct scsi_device *sdev)
{
	blk_queue_bounce_limit(sdev->request_queue, BLK_BOUNCE_ANY);
	blk_queue_dma_alignment(sdev->request_queue, 0);
	return 0;
}

static struct scsi_host_template iscsi_sht = {
	.module			= THIS_MODULE,
	.name			= "iSCSI Initiator over TCP/IP",
	.queuecommand           = iscsi_queuecommand,
	.change_queue_depth	= iscsi_change_queue_depth,
	.can_queue		= ISCSI_DEF_XMIT_CMDS_MAX - 1,
	.sg_tablesize		= 4096,
	.max_sectors		= 0xFFFF,
	.cmd_per_lun		= ISCSI_DEF_CMD_PER_LUN,
	.eh_abort_handler       = iscsi_eh_abort,
	.eh_device_reset_handler= iscsi_eh_device_reset,
	.eh_host_reset_handler	= iscsi_eh_host_reset,
	.use_clustering         = DISABLE_CLUSTERING,
	.slave_configure        = iscsi_tcp_slave_configure,
	.proc_name		= "iscsi_tcp",
	.this_id		= -1,
};

static struct iscsi_transport iscsi_tcp_transport = {
	.owner			= THIS_MODULE,
	.name			= "tcp",
	.caps			= CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST
				  | CAP_DATADGST,
	.param_mask		= ISCSI_MAX_RECV_DLENGTH |
				  ISCSI_MAX_XMIT_DLENGTH |
				  ISCSI_HDRDGST_EN |
				  ISCSI_DATADGST_EN |
				  ISCSI_INITIAL_R2T_EN |
				  ISCSI_MAX_R2T |
				  ISCSI_IMM_DATA_EN |
				  ISCSI_FIRST_BURST |
				  ISCSI_MAX_BURST |
				  ISCSI_PDU_INORDER_EN |
				  ISCSI_DATASEQ_INORDER_EN |
				  ISCSI_ERL |
				  ISCSI_CONN_PORT |
				  ISCSI_CONN_ADDRESS |
				  ISCSI_EXP_STATSN |
				  ISCSI_PERSISTENT_PORT |
				  ISCSI_PERSISTENT_ADDRESS |
				  ISCSI_TARGET_NAME | ISCSI_TPGT |
				  ISCSI_USERNAME | ISCSI_PASSWORD |
				  ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN |
				  ISCSI_FAST_ABORT | ISCSI_ABORT_TMO |
				  ISCSI_LU_RESET_TMO |
				  ISCSI_PING_TMO | ISCSI_RECV_TMO,
	.host_param_mask	= ISCSI_HOST_HWADDRESS | ISCSI_HOST_IPADDRESS |
				  ISCSI_HOST_INITIATOR_NAME |
				  ISCSI_HOST_NETDEV_NAME,
	.conndata_size		= sizeof(struct iscsi_conn),
	.sessiondata_size	= sizeof(struct iscsi_session),
	/* session management */
	.create_session		= iscsi_tcp_session_create,
	.destroy_session	= iscsi_tcp_session_destroy,
	/* connection management */
	.create_conn		= iscsi_tcp_conn_create,
	.bind_conn		= iscsi_tcp_conn_bind,
	.destroy_conn		= iscsi_tcp_conn_destroy,
	.set_param		= iscsi_conn_set_param,
	.get_conn_param		= iscsi_tcp_conn_get_param,
	.get_session_param	= iscsi_session_get_param,
	.start_conn		= iscsi_conn_start,
	.stop_conn		= iscsi_tcp_conn_stop,
	/* iscsi host params */