Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending

Pull target updates from Nicholas Bellinger:
 "There have been lots of work in a number of areas this past round.
  The highlights include:

   - Break out target_core_cdb.c emulation into SPC/SBC ops (hch)
   - Add a parse_cdb method to target backend drivers (hch)
   - Move sync_cache + write_same + unmap into spc_ops (hch)
   - Use target_execute_cmd for WRITEs in iscsi_target + srpt (hch)
   - Offload WRITE I/O backend submission in tcm_qla2xxx + tcm_fc (hch +
     nab)
   - Refactor core_update_device_list_for_node() into enable/disable
     funcs (agrover)
   - Replace the TCM processing thread with a TMR work queue (hch)
   - Fix regression in transport_add_device_to_core_hba from TMR
     conversion (DanC)
   - Remove racy, now-redundant check of sess_tearing_down with qla2xxx
     (roland)
   - Add range checking, fix reading of data len + possible underflow in
     UNMAP (roland)
   - Allow for target_submit_cmd() returning errors + convert fabrics
     (roland + nab)
   - Drop bogus struct file usage for iSCSI/SCTP (viro)"

* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (54 commits)
  iscsi-target: Drop bogus struct file usage for iSCSI/SCTP
  target: NULL dereference on error path
  target: Allow for target_submit_cmd() returning errors
  target: Check number of unmap descriptors against our limit
  target: Fix possible integer underflow in UNMAP emulation
  target: Fix reading of data length fields for UNMAP commands
  target: Add range checking to UNMAP emulation
  target: Add generation of LOGICAL BLOCK ADDRESS OUT OF RANGE
  target: Make unnecessarily global se_dev_align_max_sectors() static
  target: Remove se_session.sess_wait_list
  qla2xxx: Remove racy, now-redundant check of sess_tearing_down
  target: Check sess_tearing_down in target_get_sess_cmd()
  sbp-target: Consolidate duplicated error path code in sbp_handle_command()
  target: Un-export target_get_sess_cmd()
  qla2xxx: Get rid of redundant qla_tgt_sess.tearing_down
  target: Make core_disable_device_list_for_node use pre-refactoring lock ordering
  target: refactor core_update_device_list_for_node()
  target: Eliminate else using boolean logic
  target: Misc retval cleanups
  target: Remove hba param from core_dev_add_lun
  ...

35 files changed, 1675 insertions, 2390 deletions

diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c
index 5f6b7f63cdef..7a0ce8d42887 100644
--- a/drivers/infiniband/ulp/srpt/ib_srpt.c
+++ b/drivers/infiniband/ulp/srpt/ib_srpt.c
@@ -1377,10 +1377,14 @@ static int srpt_abort_cmd(struct srpt_send_ioctx *ioctx)
 		break;
 	case SRPT_STATE_NEED_DATA:
 		/* DMA_TO_DEVICE (write) - RDMA read error. */
+
+		/* XXX(hch): this is a horrible layering violation.. */
 		spin_lock_irqsave(&ioctx->cmd.t_state_lock, flags);
 		ioctx->cmd.transport_state |= CMD_T_LUN_STOP;
+		ioctx->cmd.transport_state &= ~CMD_T_ACTIVE;
 		spin_unlock_irqrestore(&ioctx->cmd.t_state_lock, flags);
-		transport_generic_handle_data(&ioctx->cmd);
+
+		complete(&ioctx->cmd.transport_lun_stop_comp);
 		break;
 	case SRPT_STATE_CMD_RSP_SENT:
 		/*
@@ -1463,9 +1467,10 @@ static void srpt_handle_send_comp(struct srpt_rdma_ch *ch,
 /**
  * srpt_handle_rdma_comp() - Process an IB RDMA completion notification.
  *
- * Note: transport_generic_handle_data() is asynchronous so unmapping the
- * data that has been transferred via IB RDMA must be postponed until the
- * check_stop_free() callback.
+ * XXX: what is now target_execute_cmd used to be asynchronous, and unmapping
+ * the data that has been transferred via IB RDMA had to be postponed until the
+ * check_stop_free() callback.  None of this is nessecary anymore and needs to
+ * be cleaned up.
  */
 static void srpt_handle_rdma_comp(struct srpt_rdma_ch *ch,
 				  struct srpt_send_ioctx *ioctx,
@@ -1477,7 +1482,7 @@ static void srpt_handle_rdma_comp(struct srpt_rdma_ch *ch,
 	if (opcode == SRPT_RDMA_READ_LAST) {
 		if (srpt_test_and_set_cmd_state(ioctx, SRPT_STATE_NEED_DATA,
 						SRPT_STATE_DATA_IN))
-			transport_generic_handle_data(&ioctx->cmd);
+			target_execute_cmd(&ioctx->cmd);
 		else
 			printk(KERN_ERR "%s[%d]: wrong state = %d\n", __func__,
 			       __LINE__, srpt_get_cmd_state(ioctx));
diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c
index 77759c78cc21..5b30132960c7 100644
--- a/drivers/scsi/qla2xxx/qla_target.c
+++ b/drivers/scsi/qla2xxx/qla_target.c
@@ -2643,19 +2643,9 @@ static void qlt_do_work(struct work_struct *work)
 	spin_lock_irqsave(&ha->hardware_lock, flags);
 	sess = ha->tgt.tgt_ops->find_sess_by_s_id(vha,
 	    atio->u.isp24.fcp_hdr.s_id);
-	if (sess) {
-		if (unlikely(sess->tearing_down)) {
-			sess = NULL;
-			spin_unlock_irqrestore(&ha->hardware_lock, flags);
-			goto out_term;
-		} else {
-			/*
-			 * Do the extra kref_get() before dropping
-			 * qla_hw_data->hardware_lock.
-			 */
-			kref_get(&sess->se_sess->sess_kref);
-		}
-	}
+	/* Do kref_get() before dropping qla_hw_data->hardware_lock. */
+	if (sess)
+		kref_get(&sess->se_sess->sess_kref);
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
 
 	if (unlikely(!sess)) {
diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h
index 9f9ef1644fd9..170af1571214 100644
--- a/drivers/scsi/qla2xxx/qla_target.h
+++ b/drivers/scsi/qla2xxx/qla_target.h
@@ -639,7 +639,7 @@ struct qla_tgt_func_tmpl {
 
 	int (*handle_cmd)(struct scsi_qla_host *, struct qla_tgt_cmd *,
 			unsigned char *, uint32_t, int, int, int);
-	int (*handle_data)(struct qla_tgt_cmd *);
+	void (*handle_data)(struct qla_tgt_cmd *);
 	int (*handle_tmr)(struct qla_tgt_mgmt_cmd *, uint32_t, uint8_t,
 			uint32_t);
 	void (*free_cmd)(struct qla_tgt_cmd *);
@@ -813,7 +813,6 @@ struct qla_tgt_sess {
 	unsigned int conf_compl_supported:1;
 	unsigned int deleted:1;
 	unsigned int local:1;
-	unsigned int tearing_down:1;
 
 	struct se_session *se_sess;
 	struct scsi_qla_host *vha;
diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
index 6e64314dbbb3..4752f65a9272 100644
--- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c
+++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
@@ -38,8 +38,6 @@
 #include <linux/string.h>
 #include <linux/configfs.h>
 #include <linux/ctype.h>
-#include <linux/string.h>
-#include <linux/ctype.h>
 #include <asm/unaligned.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_host.h>
@@ -466,8 +464,7 @@ static int tcm_qla2xxx_shutdown_session(struct se_session *se_sess)
 	vha = sess->vha;
 
 	spin_lock_irqsave(&vha->hw->hardware_lock, flags);
-	sess->tearing_down = 1;
-	target_splice_sess_cmd_list(se_sess);
+	target_sess_cmd_list_set_waiting(se_sess);
 	spin_unlock_irqrestore(&vha->hw->hardware_lock, flags);
 
 	return 1;
@@ -600,28 +597,15 @@ static int tcm_qla2xxx_handle_cmd(scsi_qla_host_t *vha, struct qla_tgt_cmd *cmd,
 		return -EINVAL;
 	}
 
-	target_submit_cmd(se_cmd, se_sess, cdb, &cmd->sense_buffer[0],
+	return target_submit_cmd(se_cmd, se_sess, cdb, &cmd->sense_buffer[0],
 				cmd->unpacked_lun, data_length, fcp_task_attr,
 				data_dir, flags);
-	return 0;
 }
 
-static void tcm_qla2xxx_do_rsp(struct work_struct *work)
+static void tcm_qla2xxx_handle_data_work(struct work_struct *work)
 {
 	struct qla_tgt_cmd *cmd = container_of(work, struct qla_tgt_cmd, work);
-	/*
-	 * Dispatch ->queue_status from workqueue process context
-	 */
-	transport_generic_request_failure(&cmd->se_cmd);
-}
 
-/*
- * Called from qla_target.c:qlt_do_ctio_completion()
- */
-static int tcm_qla2xxx_handle_data(struct qla_tgt_cmd *cmd)
-{
-	struct se_cmd *se_cmd = &cmd->se_cmd;
-	unsigned long flags;
 	/*
 	 * Ensure that the complete FCP WRITE payload has been received.
 	 * Otherwise return an exception via CHECK_CONDITION status.
@@ -631,24 +615,26 @@ static int tcm_qla2xxx_handle_data(struct qla_tgt_cmd *cmd)
 		 * Check if se_cmd has already been aborted via LUN_RESET, and
 		 * waiting upon completion in tcm_qla2xxx_write_pending_status()
 		 */
-		spin_lock_irqsave(&se_cmd->t_state_lock, flags);
-		if (se_cmd->transport_state & CMD_T_ABORTED) {
-			spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
-			complete(&se_cmd->t_transport_stop_comp);
-			return 0;
+		if (cmd->se_cmd.transport_state & CMD_T_ABORTED) {
+			complete(&cmd->se_cmd.t_transport_stop_comp);
+			return;
 		}
-		spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
 
-		se_cmd->scsi_sense_reason = TCM_CHECK_CONDITION_ABORT_CMD;
-		INIT_WORK(&cmd->work, tcm_qla2xxx_do_rsp);
-		queue_work(tcm_qla2xxx_free_wq, &cmd->work);
-		return 0;
+		cmd->se_cmd.scsi_sense_reason = TCM_CHECK_CONDITION_ABORT_CMD;
+		transport_generic_request_failure(&cmd->se_cmd);
+		return;
 	}
-	/*
-	 * We now tell TCM to queue this WRITE CDB with TRANSPORT_PROCESS_WRITE
-	 * status to the backstore processing thread.
-	 */
-	return transport_generic_handle_data(&cmd->se_cmd);
+
+	return target_execute_cmd(&cmd->se_cmd);
+}
+
+/*
+ * Called from qla_target.c:qlt_do_ctio_completion()
+ */
+static void tcm_qla2xxx_handle_data(struct qla_tgt_cmd *cmd)
+{
+	INIT_WORK(&cmd->work, tcm_qla2xxx_handle_data_work);
+	queue_work(tcm_qla2xxx_free_wq, &cmd->work);
 }
 
 /*
@@ -1690,7 +1676,6 @@ static struct target_core_fabric_ops tcm_qla2xxx_ops = {
 	.tpg_alloc_fabric_acl		= tcm_qla2xxx_alloc_fabric_acl,
 	.tpg_release_fabric_acl		= tcm_qla2xxx_release_fabric_acl,
 	.tpg_get_inst_index		= tcm_qla2xxx_tpg_get_inst_index,
-	.new_cmd_map			= NULL,
 	.check_stop_free		= tcm_qla2xxx_check_stop_free,
 	.release_cmd			= tcm_qla2xxx_release_cmd,
 	.put_session			= tcm_qla2xxx_put_session,
diff --git a/drivers/target/Makefile b/drivers/target/Makefile
index 61648d84fbb6..9fdcb561422f 100644
--- a/drivers/target/Makefile
+++ b/drivers/target/Makefile
@@ -9,7 +9,8 @@ target_core_mod-y		:= target_core_configfs.o \
 				   target_core_tmr.o \
 				   target_core_tpg.o \
 				   target_core_transport.o \
-				   target_core_cdb.o \
+				   target_core_sbc.o \
+				   target_core_spc.o \
 				   target_core_ua.o \
 				   target_core_rd.o \
 				   target_core_stat.o
diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index d57d10cb2e47..97c0f78c3c9c 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -429,19 +429,8 @@ int iscsit_reset_np_thread(
 
 int iscsit_del_np_comm(struct iscsi_np *np)
 {
-	if (!np->np_socket)
-		return 0;
-
-	/*
-	 * Some network transports allocate their own struct sock->file,
-	 * see  if we need to free any additional allocated resources.
-	 */
-	if (np->np_flags & NPF_SCTP_STRUCT_FILE) {
-		kfree(np->np_socket->file);
-		np->np_socket->file = NULL;
-	}
-
-	sock_release(np->np_socket);
+	if (np->np_socket)
+		sock_release(np->np_socket);
 	return 0;
 }
 
@@ -1413,8 +1402,10 @@ static int iscsit_handle_data_out(struct iscsi_conn *conn, unsigned char *buf)
 		spin_unlock_bh(&cmd->istate_lock);
 
 		iscsit_stop_dataout_timer(cmd);
-		return (!ooo_cmdsn) ? transport_generic_handle_data(
-					&cmd->se_cmd) : 0;
+		if (ooo_cmdsn)
+			return 0;
+		target_execute_cmd(&cmd->se_cmd);
+		return 0;
 	} else /* DATAOUT_CANNOT_RECOVER */
 		return -1;
 
@@ -2683,7 +2674,7 @@ static int iscsit_send_logout_response(
 		 */
 		logout_conn = iscsit_get_conn_from_cid_rcfr(sess,
 				cmd->logout_cid);
-		if ((logout_conn)) {
+		if (logout_conn) {
 			iscsit_connection_reinstatement_rcfr(logout_conn);
 			iscsit_dec_conn_usage_count(logout_conn);
 		}
@@ -4077,13 +4068,8 @@ int iscsit_close_connection(
 	kfree(conn->conn_ops);
 	conn->conn_ops = NULL;
 
-	if (conn->sock) {
-		if (conn->conn_flags & CONNFLAG_SCTP_STRUCT_FILE) {
-			kfree(conn->sock->file);
-			conn->sock->file = NULL;
-		}
+	if (conn->sock)
 		sock_release(conn->sock);
-	}
 	conn->thread_set = NULL;
 
 	pr_debug("Moving to TARG_CONN_STATE_FREE.\n");
diff --git a/drivers/target/iscsi/iscsi_target_configfs.c b/drivers/target/iscsi/iscsi_target_configfs.c
index 69dc8e35c03a..a7b25e783b58 100644
--- a/drivers/target/iscsi/iscsi_target_configfs.c
+++ b/drivers/target/iscsi/iscsi_target_configfs.c
@@ -47,28 +47,6 @@ struct lio_target_configfs_attribute {
 	ssize_t (*store)(void *, const char *, size_t);
 };
 
-struct iscsi_portal_group *lio_get_tpg_from_tpg_item(
-	struct config_item *item,
-	struct iscsi_tiqn **tiqn_out)
-{
-	struct se_portal_group *se_tpg = container_of(to_config_group(item),
-					struct se_portal_group, tpg_group);
-	struct iscsi_portal_group *tpg = se_tpg->se_tpg_fabric_ptr;
-	int ret;
-
-	if (!tpg) {
-		pr_err("Unable to locate struct iscsi_portal_group "
-			"pointer\n");
-		return NULL;
-	}
-	ret = iscsit_get_tpg(tpg);
-	if (ret < 0)
-		return NULL;
-
-	*tiqn_out = tpg->tpg_tiqn;
-	return tpg;
-}
-
 /* Start items for lio_target_portal_cit */
 
 static ssize_t lio_target_np_show_sctp(
diff --git a/drivers/target/iscsi/iscsi_target_core.h b/drivers/target/iscsi/iscsi_target_core.h
index 1c70144cdaf1..8a908b28d8b2 100644
--- a/drivers/target/iscsi/iscsi_target_core.h
+++ b/drivers/target/iscsi/iscsi_target_core.h
@@ -224,7 +224,6 @@ enum iscsi_timer_flags_table {
 /* Used for struct iscsi_np->np_flags */
 enum np_flags_table {
 	NPF_IP_NETWORK		= 0x00,
-	NPF_SCTP_STRUCT_FILE	= 0x01 /* Bugfix */
 };
 
 /* Used for struct iscsi_np->np_thread_state */
@@ -481,6 +480,7 @@ struct iscsi_tmr_req {
 	bool			task_reassign:1;
 	u32			ref_cmd_sn;
 	u32			exp_data_sn;
+	struct iscsi_cmd	*ref_cmd;
 	struct iscsi_conn_recovery *conn_recovery;
 	struct se_tmr_req	*se_tmr_req;
 };
@@ -503,7 +503,6 @@ struct iscsi_conn {
 	u16			local_port;
 	int			net_size;
 	u32			auth_id;
-#define CONNFLAG_SCTP_STRUCT_FILE			0x01
 	u32			conn_flags;
 	/* Used for iscsi_tx_login_rsp() */
 	u32			login_itt;
diff --git a/drivers/target/iscsi/iscsi_target_erl1.c b/drivers/target/iscsi/iscsi_target_erl1.c
index ecdd46deedda..3df8a2cef86f 100644
--- a/drivers/target/iscsi/iscsi_target_erl1.c
+++ b/drivers/target/iscsi/iscsi_target_erl1.c
@@ -965,8 +965,8 @@ int iscsit_execute_cmd(struct iscsi_cmd *cmd, int ooo)
 		if (cmd->immediate_data) {
 			if (cmd->cmd_flags & ICF_GOT_LAST_DATAOUT) {
 				spin_unlock_bh(&cmd->istate_lock);
-				return transport_generic_handle_data(
-						&cmd->se_cmd);
+				target_execute_cmd(&cmd->se_cmd);
+				return 0;
 			}
 			spin_unlock_bh(&cmd->istate_lock);
 
diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c
index a3656c9903a1..0694d9b1bce6 100644
--- a/drivers/target/iscsi/iscsi_target_login.c
+++ b/drivers/target/iscsi/iscsi_target_login.c
@@ -518,7 +518,7 @@ int iscsi_login_post_auth_non_zero_tsih(
 	 * initiator and release the new connection.
 	 */
 	conn_ptr = iscsit_get_conn_from_cid_rcfr(sess, cid);
-	if ((conn_ptr)) {
+	if (conn_ptr) {
 		pr_err("Connection exists with CID %hu for %s,"
 			" performing connection reinstatement.\n",
 			conn_ptr->cid, sess->sess_ops->InitiatorName);
@@ -539,7 +539,7 @@ int iscsi_login_post_auth_non_zero_tsih(
 	if (sess->sess_ops->ErrorRecoveryLevel == 2) {
 		cr = iscsit_get_inactive_connection_recovery_entry(
 				sess, cid);
-		if ((cr)) {
+		if (cr) {
 			pr_debug("Performing implicit logout"
 				" for connection recovery on CID: %hu\n",
 					conn->cid);
@@ -795,22 +795,6 @@ int iscsi_target_setup_login_socket(
 	}
 	np->np_socket = sock;
 	/*
-	 * The SCTP stack needs struct socket->file.
-	 */
-	if ((np->np_network_transport == ISCSI_SCTP_TCP) ||
-	    (np->np_network_transport == ISCSI_SCTP_UDP)) {
-		if (!sock->file) {
-			sock->file = kzalloc(sizeof(struct file), GFP_KERNEL);
-			if (!sock->file) {
-				pr_err("Unable to allocate struct"
-						" file for SCTP\n");
-				ret = -ENOMEM;
-				goto fail;
-			}
-			np->np_flags |= NPF_SCTP_STRUCT_FILE;
-		}
-	}
-	/*
 	 * Setup the np->np_sockaddr from the passed sockaddr setup
 	 * in iscsi_target_configfs.c code..
 	 */
@@ -869,21 +853,15 @@ int iscsi_target_setup_login_socket(
 
 fail:
 	np->np_socket = NULL;
-	if (sock) {
-		if (np->np_flags & NPF_SCTP_STRUCT_FILE) {
-			kfree(sock->file);
-			sock->file = NULL;
-		}
-
+	if (sock)
 		sock_release(sock);
-	}
 	return ret;
 }
 
 static int __iscsi_target_login_thread(struct iscsi_np *np)
 {
 	u8 buffer[ISCSI_HDR_LEN], iscsi_opcode, zero_tsih = 0;
-	int err, ret = 0, set_sctp_conn_flag, stop;
+	int err, ret = 0, stop;
 	struct iscsi_conn *conn = NULL;
 	struct iscsi_login *login;
 	struct iscsi_portal_group *tpg = NULL;
@@ -894,7 +872,6 @@ static int __iscsi_target_login_thread(struct iscsi_np *np)
 	struct sockaddr_in6 sock_in6;
 
 	flush_signals(current);
-	set_sctp_conn_flag = 0;
 	sock = np->np_socket;
 
 	spin_lock_bh(&np->np_thread_lock);
@@ -917,35 +894,12 @@ static int __iscsi_target_login_thread(struct iscsi_np *np)
 		spin_unlock_bh(&np->np_thread_lock);
 		goto out;
 	}
-	/*
-	 * The SCTP stack needs struct socket->file.
-	 */
-	if ((np->np_network_transport == ISCSI_SCTP_TCP) ||
-	    (np->np_network_transport == ISCSI_SCTP_UDP)) {
-		if (!new_sock->file) {
-			new_sock->file = kzalloc(
-					sizeof(struct file), GFP_KERNEL);
-			if (!new_sock->file) {
-				pr_err("Unable to allocate struct"
-						" file for SCTP\n");
-				sock_release(new_sock);
-				/* Get another socket */
-				return 1;
-			}
-			set_sctp_conn_flag = 1;
-		}
-	}
-
 	iscsi_start_login_thread_timer(np);
 
 	conn = kzalloc(sizeof(struct iscsi_conn), GFP_KERNEL);
 	if (!conn) {
 		pr_err("Could not allocate memory for"
 			" new connection\n");
-		if (set_sctp_conn_flag) {
-			kfree(new_sock->file);
-			new_sock->file = NULL;
-		}
 		sock_release(new_sock);
 		/* Get another socket */
 		return 1;
@@ -955,9 +909,6 @@ static int __iscsi_target_login_thread(struct iscsi_np *np)
 	conn->conn_state = TARG_CONN_STATE_FREE;
 	conn->sock = new_sock;
 
-	if (set_sctp_conn_flag)
-		conn->conn_flags |= CONNFLAG_SCTP_STRUCT_FILE;
-
 	pr_debug("Moving to TARG_CONN_STATE_XPT_UP.\n");
 	conn->conn_state = TARG_CONN_STATE_XPT_UP;
 
@@ -1081,7 +1032,7 @@ static int __iscsi_target_login_thread(struct iscsi_np *np)
 		goto new_sess_out;
 
 	zero_tsih = (pdu->tsih == 0x0000);
-	if ((zero_tsih)) {
+	if (zero_tsih) {
 		/*
 		 * This is the leading connection of a new session.
 		 * We wait until after authentication to check for
@@ -1205,13 +1156,8 @@ old_sess_out:
 		iscsi_release_param_list(conn->param_list);
 		conn->param_list = NULL;
 	}
-	if (conn->sock) {
-		if (conn->conn_flags & CONNFLAG_SCTP_STRUCT_FILE) {
-			kfree(conn->sock->file);
-			conn->sock->file = NULL;
-		}
+	if (conn->sock)
 		sock_release(conn->sock);
-	}
 	kfree(conn);
 
 	if (tpg) {
diff --git a/drivers/target/iscsi/iscsi_target_parameters.c b/drivers/target/iscsi/iscsi_target_parameters.c
index ed5241e7f12a..0c4760fabfc0 100644
--- a/drivers/target/iscsi/iscsi_target_parameters.c
+++ b/drivers/target/iscsi/iscsi_target_parameters.c
@@ -681,7 +681,7 @@ int iscsi_update_param_value(struct iscsi_param *param, char *value)
 	param->value = kzalloc(strlen(value) + 1, GFP_KERNEL);
 	if (!param->value) {
 		pr_err("Unable to allocate memory for value.\n");
-		return -1;
+		return -ENOMEM;
 	}
 
 	memcpy(param->value, value, strlen(value));
diff --git a/drivers/target/iscsi/iscsi_target_tmr.c b/drivers/target/iscsi/iscsi_target_tmr.c
index f4e640b51fd1..f62fe123d902 100644
--- a/drivers/target/iscsi/iscsi_target_tmr.c
+++ b/drivers/target/iscsi/iscsi_target_tmr.c
@@ -19,6 +19,7 @@
  ******************************************************************************/
 
 #include <asm/unaligned.h>
+#include <scsi/scsi_device.h>
 #include <scsi/iscsi_proto.h>
 #include <target/target_core_base.h>
 #include <target/target_core_fabric.h>
@@ -61,7 +62,7 @@ u8 iscsit_tmr_abort_task(
 	}
 
 	se_tmr->ref_task_tag		= hdr->rtt;
-	se_tmr->ref_cmd			= &ref_cmd->se_cmd;
+	tmr_req->ref_cmd		= ref_cmd;
 	tmr_req->ref_cmd_sn		= hdr->refcmdsn;
 	tmr_req->exp_data_sn		= hdr->exp_datasn;
 
@@ -121,7 +122,7 @@ u8 iscsit_tmr_task_reassign(
 	struct iscsi_tmr_req *tmr_req = cmd->tmr_req;
 	struct se_tmr_req *se_tmr = cmd->se_cmd.se_tmr_req;
 	struct iscsi_tm *hdr = (struct iscsi_tm *) buf;
-	int ret;
+	int ret, ref_lun;
 
 	pr_debug("Got TASK_REASSIGN TMR ITT: 0x%08x,"
 		" RefTaskTag: 0x%08x, ExpDataSN: 0x%08x, CID: %hu\n",
@@ -155,9 +156,16 @@ u8 iscsit_tmr_task_reassign(
 		return ISCSI_TMF_RSP_REJECTED;
 	}
 
+	ref_lun = scsilun_to_int(&hdr->lun);
+	if (ref_lun != ref_cmd->se_cmd.orig_fe_lun) {
+		pr_err("Unable to perform connection recovery for"
+			" differing ref_lun: %d ref_cmd orig_fe_lun: %u\n",
+			ref_lun, ref_cmd->se_cmd.orig_fe_lun);
+		return ISCSI_TMF_RSP_REJECTED;
+	}
+
 	se_tmr->ref_task_tag		= hdr->rtt;
-	se_tmr->ref_cmd			= &ref_cmd->se_cmd;
-	se_tmr->ref_task_lun		= get_unaligned_le64(&hdr->lun);
+	tmr_req->ref_cmd		= ref_cmd;
 	tmr_req->ref_cmd_sn		= hdr->refcmdsn;
 	tmr_req->exp_data_sn		= hdr->exp_datasn;
 	tmr_req->conn_recovery		= cr;
@@ -191,9 +199,7 @@ static int iscsit_task_reassign_complete_nop_out(
 	struct iscsi_tmr_req *tmr_req,
 	struct iscsi_conn *conn)
 {
-	struct se_tmr_req *se_tmr = tmr_req->se_tmr_req;
-	struct se_cmd *se_cmd = se_tmr->ref_cmd;
-	struct iscsi_cmd *cmd = container_of(se_cmd, struct iscsi_cmd, se_cmd);
+	struct iscsi_cmd *cmd = tmr_req->ref_cmd;
 	struct iscsi_conn_recovery *cr;
 
 	if (!cmd->cr) {
@@ -251,7 +257,8 @@ static int iscsit_task_reassign_complete_write(
 			pr_debug("WRITE ITT: 0x%08x: t_state: %d"
 				" never sent to transport\n",
 				cmd->init_task_tag, cmd->se_cmd.t_state);
-			return transport_generic_handle_data(se_cmd);
+			target_execute_cmd(se_cmd);
+			return 0;
 		}
 
 		cmd->i_state = ISTATE_SEND_STATUS;
@@ -360,9 +367,7 @@ static int iscsit_task_reassign_complete_scsi_cmnd(
 	struct iscsi_tmr_req *tmr_req,
 	struct iscsi_conn *conn)
 {
-	struct se_tmr_req *se_tmr = tmr_req->se_tmr_req;
-	struct se_cmd *se_cmd = se_tmr->ref_cmd;
-	struct iscsi_cmd *cmd = container_of(se_cmd, struct iscsi_cmd, se_cmd);
+	struct iscsi_cmd *cmd = tmr_req->ref_cmd;
 	struct iscsi_conn_recovery *cr;
 
 	if (!cmd->cr) {
@@ -385,7 +390,7 @@ static int iscsit_task_reassign_complete_scsi_cmnd(
 	list_add_tail(&cmd->i_conn_node, &conn->conn_cmd_list);
 	spin_unlock_bh(&conn->cmd_lock);
 
-	if (se_cmd->se_cmd_flags & SCF_SENT_CHECK_CONDITION) {
+	if (cmd->se_cmd.se_cmd_flags & SCF_SENT_CHECK_CONDITION) {
 		cmd->i_state = ISTATE_SEND_STATUS;
 		iscsit_add_cmd_to_response_queue(cmd, conn, cmd->i_state);
 		return 0;
@@ -411,17 +416,14 @@ static int iscsit_task_reassign_complete(
 	struct iscsi_tmr_req *tmr_req,
 	struct iscsi_conn *conn)
 {
-	struct se_tmr_req *se_tmr = tmr_req->se_tmr_req;
-	struct se_cmd *se_cmd;
 	struct iscsi_cmd *cmd;
 	int ret = 0;
 
-	if (!se_tmr->ref_cmd) {
+	if (!tmr_req->ref_cmd) {
 		pr_err("TMR Request is missing a RefCmd struct iscsi_cmd.\n");
 		return -1;
 	}
-	se_cmd = se_tmr->ref_cmd;
-	cmd = container_of(se_cmd, struct iscsi_cmd, se_cmd);
+	cmd = tmr_req->ref_cmd;
 
 	cmd->conn = conn;
 
@@ -547,9 +549,7 @@ int iscsit_task_reassign_prepare_write(
 	struct iscsi_tmr_req *tmr_req,
 	struct iscsi_conn *conn)
 {
-	struct se_tmr_req *se_tmr = tmr_req->se_tmr_req;
-	struct se_cmd *se_cmd = se_tmr->ref_cmd;
-	struct iscsi_cmd *cmd = container_of(se_cmd, struct iscsi_cmd, se_cmd);
+	struct iscsi_cmd *cmd = tmr_req->ref_cmd;
 	struct iscsi_pdu *pdu = NULL;
 	struct iscsi_r2t *r2t = NULL, *r2t_tmp;
 	int first_incomplete_r2t = 1, i = 0;
@@ -782,14 +782,12 @@ int iscsit_check_task_reassign_expdatasn(
 	struct iscsi_tmr_req *tmr_req,
 	struct iscsi_conn *conn)
 {
-	struct se_tmr_req *se_tmr = tmr_req->se_tmr_req;
-	struct se_cmd *se_cmd = se_tmr->ref_cmd;
-	struct iscsi_cmd *ref_cmd = container_of(se_cmd, struct iscsi_cmd, se_cmd);
+	struct iscsi_cmd *ref_cmd = tmr_req->ref_cmd;
 
 	if (ref_cmd->iscsi_opcode != ISCSI_OP_SCSI_CMD)
 		return 0;
 
-	if (se_cmd->se_cmd_flags & SCF_SENT_CHECK_CONDITION)
+	if (ref_cmd->se_cmd.se_cmd_flags & SCF_SENT_CHECK_CONDITION)
 		return 0;
 
 	if (ref_cmd->data_direction == DMA_NONE)
diff --git a/drivers/target/iscsi/iscsi_target_tpg.c b/drivers/target/iscsi/iscsi_target_tpg.c
index 879d8d0fa3fe..a38a3f8ab0d9 100644
--- a/drivers/target/iscsi/iscsi_target_tpg.c
+++ b/drivers/target/iscsi/iscsi_target_tpg.c
@@ -303,6 +303,7 @@ int iscsit_tpg_enable_portal_group(struct iscsi_portal_group *tpg)
 {
 	struct iscsi_param *param;
 	struct iscsi_tiqn *tiqn = tpg->tpg_tiqn;
+	int ret;
 
 	spin_lock(&tpg->tpg_state_lock);
 	if (tpg->tpg_state == TPG_STATE_ACTIVE) {
@@ -319,19 +320,19 @@ int iscsit_tpg_enable_portal_group(struct iscsi_portal_group *tpg)
 	param = iscsi_find_param_from_key(AUTHMETHOD, tpg->param_list);
 	if (!param) {
 		spin_unlock(&tpg->tpg_state_lock);
-		return -ENOMEM;
+		return -EINVAL;
 	}
 
 	if (ISCSI_TPG_ATTRIB(tpg)->authentication) {
-		if (!strcmp(param->value, NONE))
-			if (iscsi_update_param_value(param, CHAP) < 0) {
-				spin_unlock(&tpg->tpg_state_lock);
-				return -ENOMEM;
-			}
-		if (iscsit_ta_authentication(tpg, 1) < 0) {
-			spin_unlock(&tpg->tpg_state_lock);
-			return -ENOMEM;
+		if (!strcmp(param->value, NONE)) {
+			ret = iscsi_update_param_value(param, CHAP);
+			if (ret)
+				goto err;
 		}
+
+		ret = iscsit_ta_authentication(tpg, 1);
+		if (ret < 0)
+			goto err;
 	}
 
 	tpg->tpg_state = TPG_STATE_ACTIVE;
@@ -344,6 +345,10 @@ int iscsit_tpg_enable_portal_group(struct iscsi_portal_group *tpg)
 	spin_unlock(&tiqn->tiqn_tpg_lock);
 
 	return 0;
+
+err:
+	spin_unlock(&tpg->tpg_state_lock);
+	return ret;
 }
 
 int iscsit_tpg_disable_portal_group(struct iscsi_portal_group *tpg, int force)
@@ -558,7 +563,7 @@ int iscsit_ta_authentication(struct iscsi_portal_group *tpg, u32 authentication)
 	if ((authentication != 1) && (authentication != 0)) {
 		pr_err("Illegal value for authentication parameter:"
 			" %u, ignoring request.\n", authentication);
-		return -1;
+		return -EINVAL;
 	}
 
 	memset(buf1, 0, sizeof(buf1));
@@ -593,7 +598,7 @@ int iscsit_ta_authentication(struct iscsi_portal_group *tpg, u32 authentication)
 	} else {
 		snprintf(buf1, sizeof(buf1), "%s", param->value);
 		none = strstr(buf1, NONE);
-		if ((none))
+		if (none)
 			goto out;
 		strncat(buf1, ",", strlen(","));
 		strncat(buf1, NONE, strlen(NONE));
diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c
index 38dfac2b0a1c..5491c632a15e 100644
--- a/drivers/target/loopback/tcm_loop.c
+++ b/drivers/target/loopback/tcm_loop.c
@@ -211,12 +211,11 @@ static void tcm_loop_submission_work(struct work_struct *work)
 	/*
 	 * Because some userspace code via scsi-generic do not memset their
 	 * associated read buffers, go ahead and do that here for type
-	 * SCF_SCSI_CONTROL_SG_IO_CDB.  Also note that this is currently
-	 * guaranteed to be a single SGL for SCF_SCSI_CONTROL_SG_IO_CDB
-	 * by target core in target_setup_cmd_from_cdb() ->
-	 * transport_generic_cmd_sequencer().
+	 * non-data CDBs.  Also note that this is currently guaranteed to be a
+	 * single SGL for this case by target core in
+	 * target_setup_cmd_from_cdb() -> transport_generic_cmd_sequencer().
 	 */
-	if (se_cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB &&
+	if (!(se_cmd->se_cmd_flags & SCF_SCSI_DATA_CDB) &&
 	    se_cmd->data_direction == DMA_FROM_DEVICE) {
 		struct scatterlist *sg = scsi_sglist(sc);
 		unsigned char *buf = kmap(sg_page(sg)) + sg->offset;
@@ -779,7 +778,7 @@ static int tcm_loop_write_pending(struct se_cmd *se_cmd)
 	 * We now tell TCM to add this WRITE CDB directly into the TCM storage
 	 * object execution queue.
 	 */
-	transport_generic_process_write(se_cmd);
+	target_execute_cmd(se_cmd);
 	return 0;
 }
 
diff --git a/drivers/target/sbp/sbp_target.c b/drivers/target/sbp/sbp_target.c
index 7e6136e2ce81..39ddba584b30 100644
--- a/drivers/target/sbp/sbp_target.c
+++ b/drivers/target/sbp/sbp_target.c
@@ -1219,28 +1219,14 @@ static void sbp_handle_command(struct sbp_target_request *req)
 	ret = sbp_fetch_command(req);
 	if (ret) {
 		pr_debug("sbp_handle_command: fetch command failed: %d\n", ret);
-		req->status.status |= cpu_to_be32(
-			STATUS_BLOCK_RESP(STATUS_RESP_TRANSPORT_FAILURE) |
-			STATUS_BLOCK_DEAD(0) |
-			STATUS_BLOCK_LEN(1) |
-			STATUS_BLOCK_SBP_STATUS(SBP_STATUS_UNSPECIFIED_ERROR));
-		sbp_send_status(req);
-		sbp_free_request(req);
-		return;
+		goto err;
 	}
 
 	ret = sbp_fetch_page_table(req);
 	if (ret) {
 		pr_debug("sbp_handle_command: fetch page table failed: %d\n",
 			ret);
-		req->status.status |= cpu_to_be32(
-			STATUS_BLOCK_RESP(STATUS_RESP_TRANSPORT_FAILURE) |
-			STATUS_BLOCK_DEAD(0) |
-			STATUS_BLOCK_LEN(1) |
-			STATUS_BLOCK_SBP_STATUS(SBP_STATUS_UNSPECIFIED_ERROR));
-		sbp_send_status(req);
-		sbp_free_request(req);
-		return;
+		goto err;
 	}
 
 	unpacked_lun = req->login->lun->unpacked_lun;
@@ -1249,9 +1235,21 @@ static void sbp_handle_command(struct sbp_target_request *req)
 	pr_debug("sbp_handle_command ORB:0x%llx unpacked_lun:%d data_len:%d data_dir:%d\n",
 			req->orb_pointer, unpacked_lun, data_length, data_dir);
 
-	target_submit_cmd(&req->se_cmd, sess->se_sess, req->cmd_buf,
-			req->sense_buf, unpacked_lun, data_length,
-			MSG_SIMPLE_TAG, data_dir, 0);
+	if (target_submit_cmd(&req->se_cmd, sess->se_sess, req->cmd_buf,
+			      req->sense_buf, unpacked_lun, data_length,
+			      MSG_SIMPLE_TAG, data_dir, 0))
+		goto err;
+
+	return;
+
+err:
+	req->status.status |= cpu_to_be32(
+		STATUS_BLOCK_RESP(STATUS_RESP_TRANSPORT_FAILURE) |
+		STATUS_BLOCK_DEAD(0) |
+		STATUS_BLOCK_LEN(1) |
+		STATUS_BLOCK_SBP_STATUS(SBP_STATUS_UNSPECIFIED_ERROR));
+	sbp_send_status(req);
+	sbp_free_request(req);
 }
 
 /*
@@ -1784,8 +1782,7 @@ static int sbp_write_pending(struct se_cmd *se_cmd)
 		return ret;
 	}
 
-	transport_generic_process_write(se_cmd);
-
+	target_execute_cmd(se_cmd);
 	return 0;
 }
 
diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index 5ad972856a8d..cf2c66f3c116 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -300,8 +300,8 @@ int core_free_device_list_for_node(
 		lun = deve->se_lun;
 
 		spin_unlock_irq(&nacl->device_list_lock);
-		core_update_device_list_for_node(lun, NULL, deve->mapped_lun,
-			TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg, 0);
+		core_disable_device_list_for_node(lun, NULL, deve->mapped_lun,
+			TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg);
 		spin_lock_irq(&nacl->device_list_lock);
 	}
 	spin_unlock_irq(&nacl->device_list_lock);
@@ -342,72 +342,46 @@ void core_update_device_list_access(
 	spin_unlock_irq(&nacl->device_list_lock);
 }
 
-/*      core_update_device_list_for_node():
+/*      core_enable_device_list_for_node():
  *
  *
  */
-int core_update_device_list_for_node(
+int core_enable_device_list_for_node(
 	struct se_lun *lun,
 	struct se_lun_acl *lun_acl,
 	u32 mapped_lun,
 	u32 lun_access,
 	struct se_node_acl *nacl,
-	struct se_portal_group *tpg,
-	int enable)
+	struct se_portal_group *tpg)
 {
 	struct se_port *port = lun->lun_sep;
-	struct se_dev_entry *deve = nacl->device_list[mapped_lun];
-	int trans = 0;
-	/*
-	 * If the MappedLUN entry is being disabled, the entry in
-	 * port->sep_alua_list must be removed now before clearing the
-	 * struct se_dev_entry pointers below as logic in
-	 * core_alua_do_transition_tg_pt() depends on these being present.
-	 */
-	if (!enable) {
-		/*
-		 * deve->se_lun_acl will be NULL for demo-mode created LUNs
-		 * that have not been explicitly concerted to MappedLUNs ->
-		 * struct se_lun_acl, but we remove deve->alua_port_list from
-		 * port->sep_alua_list. This also means that active UAs and
-		 * NodeACL context specific PR metadata for demo-mode
-		 * MappedLUN *deve will be released below..
-		 */
-		spin_lock_bh(&port->sep_alua_lock);
-		list_del(&deve->alua_port_list);
-		spin_unlock_bh(&port->sep_alua_lock);
-	}
+	struct se_dev_entry *deve;
 
 	spin_lock_irq(&nacl->device_list_lock);
-	if (enable) {
-		/*
-		 * Check if the call is handling demo mode -> explict LUN ACL
-		 * transition.  This transition must be for the same struct se_lun
-		 * + mapped_lun that was setup in demo mode..
-		 */
-		if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) {
-			if (deve->se_lun_acl != NULL) {
-				pr_err("struct se_dev_entry->se_lun_acl"
-					" already set for demo mode -> explict"
-					" LUN ACL transition\n");
-				spin_unlock_irq(&nacl->device_list_lock);
-				return -EINVAL;
-			}
-			if (deve->se_lun != lun) {
-				pr_err("struct se_dev_entry->se_lun does"
-					" match passed struct se_lun for demo mode"
-					" -> explict LUN ACL transition\n");
-				spin_unlock_irq(&nacl->device_list_lock);
-				return -EINVAL;
-			}
-			deve->se_lun_acl = lun_acl;
-			trans = 1;
-		} else {
-			deve->se_lun = lun;
-			deve->se_lun_acl = lun_acl;
-			deve->mapped_lun = mapped_lun;
-			deve->lun_flags |= TRANSPORT_LUNFLAGS_INITIATOR_ACCESS;
+
+	deve = nacl->device_list[mapped_lun];
+
+	/*
+	 * Check if the call is handling demo mode -> explict LUN ACL
+	 * transition.  This transition must be for the same struct se_lun
+	 * + mapped_lun that was setup in demo mode..
+	 */
+	if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) {
+		if (deve->se_lun_acl != NULL) {
+			pr_err("struct se_dev_entry->se_lun_acl"
+			       " already set for demo mode -> explict"
+			       " LUN ACL transition\n");
+			spin_unlock_irq(&nacl->device_list_lock);
+			return -EINVAL;
 		}
+		if (deve->se_lun != lun) {
+			pr_err("struct se_dev_entry->se_lun does"
+			       " match passed struct se_lun for demo mode"
+			       " -> explict LUN ACL transition\n");
+			spin_unlock_irq(&nacl->device_list_lock);
+			return -EINVAL;
+		}
+		deve->se_lun_acl = lun_acl;
 
 		if (lun_access & TRANSPORT_LUNFLAGS_READ_WRITE) {
 			deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_ONLY;
@@ -417,27 +391,72 @@ int core_update_device_list_for_node(
 			deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_ONLY;
 		}
 
-		if (trans) {
-			spin_unlock_irq(&nacl->device_list_lock);
-			return 0;
-		}
-		deve->creation_time = get_jiffies_64();
-		deve->attach_count++;
 		spin_unlock_irq(&nacl->device_list_lock);
+		return 0;
+	}
 
-		spin_lock_bh(&port->sep_alua_lock);
-		list_add_tail(&deve->alua_port_list, &port->sep_alua_list);
-		spin_unlock_bh(&port->sep_alua_lock);
+	deve->se_lun = lun;
+	deve->se_lun_acl = lun_acl;
+	deve->mapped_lun = mapped_lun;
+	deve->lun_flags |= TRANSPORT_LUNFLAGS_INITIATOR_ACCESS;
 
-		return 0;
+	if (lun_access & TRANSPORT_LUNFLAGS_READ_WRITE) {
+		deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_ONLY;
+		deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_WRITE;
+	} else {
+		deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_WRITE;
+		deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_ONLY;
 	}
+
+	deve->creation_time = get_jiffies_64();
+	deve->attach_count++;
+	spin_unlock_irq(&nacl->device_list_lock);
+
+	spin_lock_bh(&port->sep_alua_lock);
+	list_add_tail(&deve->alua_port_list, &port->sep_alua_list);
+	spin_unlock_bh(&port->sep_alua_lock);
+
+	return 0;
+}
+
+/*      core_disable_device_list_for_node():
+ *
+ *
+ */
+int core_disable_device_list_for_node(
+	struct se_lun *lun,
+	struct se_lun_acl *lun_acl,
+	u32 mapped_lun,
+	u32 lun_access,
+	struct se_node_acl *nacl,
+	struct se_portal_group *tpg)
+{
+	struct se_port *port = lun->lun_sep;
+	struct se_dev_entry *deve = nacl->device_list[mapped_lun];
+
+	/*
+	 * If the MappedLUN entry is being disabled, the entry in
+	 * port->sep_alua_list must be removed now before clearing the
+	 * struct se_dev_entry pointers below as logic in
+	 * core_alua_do_transition_tg_pt() depends on these being present.
+	 *
+	 * deve->se_lun_acl will be NULL for demo-mode created LUNs
+	 * that have not been explicitly converted to MappedLUNs ->
+	 * struct se_lun_acl, but we remove deve->alua_port_list from
+	 * port->sep_alua_list. This also means that active UAs and
+	 * NodeACL context specific PR metadata for demo-mode
+	 * MappedLUN *deve will be released below..
+	 */
+	spin_lock_bh(&port->sep_alua_lock);
+	list_del(&deve->alua_port_list);
+	spin_unlock_bh(&port->sep_alua_lock);
 	/*
 	 * Wait for any in process SPEC_I_PT=1 or REGISTER_AND_MOVE
 	 * PR operation to complete.
 	 */
-	spin_unlock_irq(&nacl->device_list_lock);
 	while (atomic_read(&deve->pr_ref_count) != 0)
 		cpu_relax();
+
 	spin_lock_irq(&nacl->device_list_lock);
 	/*
 	 * Disable struct se_dev_entry LUN ACL mapping
@@ -475,9 +494,9 @@ void core_clear_lun_from_tpg(struct se_lun *lun, struct se_portal_group *tpg)
 				continue;
 			spin_unlock_irq(&nacl->device_list_lock);
 
-			core_update_device_list_for_node(lun, NULL,
+			core_disable_device_list_for_node(lun, NULL,
 				deve->mapped_lun, TRANSPORT_LUNFLAGS_NO_ACCESS,
-				nacl, tpg, 0);
+				nacl, tpg);
 
 			spin_lock_irq(&nacl->device_list_lock);
 		}
@@ -715,7 +734,7 @@ void se_release_device_for_hba(struct se_device *dev)
 		se_dev_stop(dev);
 
 	if (dev->dev_ptr) {
-		kthread_stop(dev->process_thread);
+		destroy_workqueue(dev->tmr_wq);
 		if (dev->transport->free_device)
 			dev->transport->free_device(dev->dev_ptr);
 	}
@@ -822,7 +841,7 @@ int se_dev_check_shutdown(struct se_device *dev)
 	return ret;
 }
 
-u32 se_dev_align_max_sectors(u32 max_sectors, u32 block_size)
+static u32 se_dev_align_max_sectors(u32 max_sectors, u32 block_size)
 {
 	u32 tmp, aligned_max_sectors;
 	/*
@@ -1273,7 +1292,6 @@ int se_dev_set_block_size(struct se_device *dev, u32 block_size)
 
 struct se_lun *core_dev_add_lun(
 	struct se_portal_group *tpg,
-	struct se_hba *hba,
 	struct se_device *dev,
 	u32 lun)
 {
@@ -1298,7 +1316,7 @@ struct se_lun *core_dev_add_lun(
 	pr_debug("%s_TPG[%u]_LUN[%u] - Activated %s Logical Unit from"
 		" CORE HBA: %u\n", tpg->se_tpg_tfo->get_fabric_name(),
 		tpg->se_tpg_tfo->tpg_get_tag(tpg), lun_p->unpacked_lun,
-		tpg->se_tpg_tfo->get_fabric_name(), hba->hba_id);
+		tpg->se_tpg_tfo->get_fabric_name(), dev->se_hba->hba_id);
 	/*
 	 * Update LUN maps for dynamically added initiators when
 	 * generate_node_acl is enabled.
@@ -1470,8 +1488,8 @@ int core_dev_add_initiator_node_lun_acl(
 
 	lacl->se_lun = lun;
 
-	if (core_update_device_list_for_node(lun, lacl, lacl->mapped_lun,
-			lun_access, nacl, tpg, 1) < 0)
+	if (core_enable_device_list_for_node(lun, lacl, lacl->mapped_lun,
+			lun_access, nacl, tpg) < 0)
 		return -EINVAL;
 
 	spin_lock(&lun->lun_acl_lock);
@@ -1514,8 +1532,8 @@ int core_dev_del_initiator_node_lun_acl(
 	smp_mb__after_atomic_dec();
 	spin_unlock(&lun->lun_acl_lock);
 
-	core_update_device_list_for_node(lun, NULL, lacl->mapped_lun,
-		TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg, 0);
+	core_disable_device_list_for_node(lun, NULL, lacl->mapped_lun,
+		TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg);
 
 	lacl->se_lun = NULL;
 
diff --git a/drivers/target/target_core_fabric_configfs.c b/drivers/target/target_core_fabric_configfs.c
index 405cc98eaed6..ea479e54f5fd 100644
--- a/drivers/target/target_core_fabric_configfs.c
+++ b/drivers/target/target_core_fabric_configfs.c
@@ -764,8 +764,7 @@ static int target_fabric_port_link(
 		goto out;
 	}
 
-	lun_p = core_dev_add_lun(se_tpg, dev->se_hba, dev,
-				lun->unpacked_lun);
+	lun_p = core_dev_add_lun(se_tpg, dev, lun->unpacked_lun);
 	if (IS_ERR(lun_p)) {
 		pr_err("core_dev_add_lun() failed\n");
 		ret = PTR_ERR(lun_p);
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index 9f99d0404908..9e2100551c78 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -331,7 +331,7 @@ static int fd_do_writev(struct se_cmd *cmd, struct scatterlist *sgl,
 	return 1;
 }
 
-static void fd_emulate_sync_cache(struct se_cmd *cmd)
+static int fd_execute_sync_cache(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
 	struct fd_dev *fd_dev = dev->dev_ptr;
@@ -365,7 +365,7 @@ static void fd_emulate_sync_cache(struct se_cmd *cmd)
 		pr_err("FILEIO: vfs_fsync_range() failed: %d\n", ret);
 
 	if (immed)
-		return;
+		return 0;
 
 	if (ret) {
 		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
@@ -373,11 +373,15 @@ static void fd_emulate_sync_cache(struct se_cmd *cmd)
 	} else {
 		target_complete_cmd(cmd, SAM_STAT_GOOD);
 	}
+
+	return 0;
 }
 
-static int fd_execute_cmd(struct se_cmd *cmd, struct scatterlist *sgl,
-		u32 sgl_nents, enum dma_data_direction data_direction)
+static int fd_execute_rw(struct se_cmd *cmd)
 {
+	struct scatterlist *sgl = cmd->t_data_sg;
+	u32 sgl_nents = cmd->t_data_nents;
+	enum dma_data_direction data_direction = cmd->data_direction;
 	struct se_device *dev = cmd->se_dev;
 	int ret = 0;
 
@@ -550,6 +554,16 @@ static sector_t fd_get_blocks(struct se_device *dev)
 	return div_u64(dev_size, dev->se_sub_dev->se_dev_attrib.block_size);
 }
 
+static struct spc_ops fd_spc_ops = {
+	.execute_rw		= fd_execute_rw,
+	.execute_sync_cache	= fd_execute_sync_cache,
+};
+
+static int fd_parse_cdb(struct se_cmd *cmd)
+{
+	return sbc_parse_cdb(cmd, &fd_spc_ops);
+}
+
 static struct se_subsystem_api fileio_template = {
 	.name			= "fileio",
 	.owner			= THIS_MODULE,
@@ -561,8 +575,7 @@ static struct se_subsystem_api fileio_template = {
 	.allocate_virtdevice	= fd_allocate_virtdevice,
 	.create_virtdevice	= fd_create_virtdevice,
 	.free_device		= fd_free_device,
-	.execute_cmd		= fd_execute_cmd,
-	.do_sync_cache		= fd_emulate_sync_cache,
+	.parse_cdb		= fd_parse_cdb,
 	.check_configfs_dev_params = fd_check_configfs_dev_params,
 	.set_configfs_dev_params = fd_set_configfs_dev_params,
 	.show_configfs_dev_params = fd_show_configfs_dev_params,
diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c
index fd47950727b4..76db75e836ed 100644
--- a/drivers/target/target_core_iblock.c
+++ b/drivers/target/target_core_iblock.c
@@ -40,6 +40,7 @@
 #include <linux/module.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_host.h>
+#include <asm/unaligned.h>
 
 #include <target/target_core_base.h>
 #include <target/target_core_backend.h>
@@ -96,6 +97,7 @@ static struct se_device *iblock_create_virtdevice(
 	struct request_queue *q;
 	struct queue_limits *limits;
 	u32 dev_flags = 0;
+	fmode_t mode;
 	int ret = -EINVAL;
 
 	if (!ib_dev) {
@@ -117,8 +119,11 @@ static struct se_device *iblock_create_virtdevice(
 	pr_debug( "IBLOCK: Claiming struct block_device: %s\n",
 			ib_dev->ibd_udev_path);
 
-	bd = blkdev_get_by_path(ib_dev->ibd_udev_path,
-				FMODE_WRITE|FMODE_READ|FMODE_EXCL, ib_dev);
+	mode = FMODE_READ|FMODE_EXCL;
+	if (!ib_dev->ibd_readonly)
+		mode |= FMODE_WRITE;
+
+	bd = blkdev_get_by_path(ib_dev->ibd_udev_path, mode, ib_dev);
 	if (IS_ERR(bd)) {
 		ret = PTR_ERR(bd);
 		goto failed;
@@ -292,7 +297,7 @@ static void iblock_end_io_flush(struct bio *bio, int err)
  * Implement SYCHRONIZE CACHE.  Note that we can't handle lba ranges and must
  * always flush the whole cache.
  */
-static void iblock_emulate_sync_cache(struct se_cmd *cmd)
+static int iblock_execute_sync_cache(struct se_cmd *cmd)
 {
 	struct iblock_dev *ib_dev = cmd->se_dev->dev_ptr;
 	int immed = (cmd->t_task_cdb[1] & 0x2);
@@ -311,23 +316,98 @@ static void iblock_emulate_sync_cache(struct se_cmd *cmd)
 	if (!immed)
 		bio->bi_private = cmd;
 	submit_bio(WRITE_FLUSH, bio);
+	return 0;
 }
 
-static int iblock_do_discard(struct se_device *dev, sector_t lba, u32 range)
+static int iblock_execute_unmap(struct se_cmd *cmd)
 {
+	struct se_device *dev = cmd->se_dev;
 	struct iblock_dev *ibd = dev->dev_ptr;
-	struct block_device *bd = ibd->ibd_bd;
-	int barrier = 0;
+	unsigned char *buf, *ptr = NULL;
+	sector_t lba;
+	int size = cmd->data_length;
+	u32 range;
+	int ret = 0;
+	int dl, bd_dl;
+
+	buf = transport_kmap_data_sg(cmd);
+
+	dl = get_unaligned_be16(&buf[0]);
+	bd_dl = get_unaligned_be16(&buf[2]);
+
+	size = min(size - 8, bd_dl);
+	if (size / 16 > dev->se_sub_dev->se_dev_attrib.max_unmap_block_desc_count) {
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
+		goto err;
+	}
+
+	/* First UNMAP block descriptor starts at 8 byte offset */
+	ptr = &buf[8];
+	pr_debug("UNMAP: Sub: %s Using dl: %u bd_dl: %u size: %u"
+		" ptr: %p\n", dev->transport->name, dl, bd_dl, size, ptr);
+
+	while (size >= 16) {
+		lba = get_unaligned_be64(&ptr[0]);
+		range = get_unaligned_be32(&ptr[8]);
+		pr_debug("UNMAP: Using lba: %llu and range: %u\n",
+				 (unsigned long long)lba, range);
+
+		if (range > dev->se_sub_dev->se_dev_attrib.max_unmap_lba_count) {
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			ret = -EINVAL;
+			goto err;
+		}
+
+		if (lba + range > dev->transport->get_blocks(dev) + 1) {
+			cmd->scsi_sense_reason = TCM_ADDRESS_OUT_OF_RANGE;
+			ret = -EINVAL;
+			goto err;
+		}
 
-	return blkdev_issue_discard(bd, lba, range, GFP_KERNEL, barrier);
+		ret = blkdev_issue_discard(ibd->ibd_bd, lba, range,
+					   GFP_KERNEL, 0);
+		if (ret < 0) {
+			pr_err("blkdev_issue_discard() failed: %d\n",
+					ret);
+			goto err;
+		}
+
+		ptr += 16;
+		size -= 16;
+	}
+
+err:
+	transport_kunmap_data_sg(cmd);
+	if (!ret)
+		target_complete_cmd(cmd, GOOD);
+	return ret;
+}
+
+static int iblock_execute_write_same(struct se_cmd *cmd)
+{
+	struct iblock_dev *ibd = cmd->se_dev->dev_ptr;
+	int ret;
+
+	ret = blkdev_issue_discard(ibd->ibd_bd, cmd->t_task_lba,
+				   spc_get_write_same_sectors(cmd), GFP_KERNEL,
+				   0);
+	if (ret < 0) {
+		pr_debug("blkdev_issue_discard() failed for WRITE_SAME\n");
+		return ret;
+	}
+
+	target_complete_cmd(cmd, GOOD);
+	return 0;
 }
 
 enum {
-	Opt_udev_path, Opt_force, Opt_err
+	Opt_udev_path, Opt_readonly, Opt_force, Opt_err
 };
 
 static match_table_t tokens = {
 	{Opt_udev_path, "udev_path=%s"},
+	{Opt_readonly, "readonly=%d"},
 	{Opt_force, "force=%d"},
 	{Opt_err, NULL}
 };
@@ -340,6 +420,7 @@ static ssize_t iblock_set_configfs_dev_params(struct se_hba *hba,
 	char *orig, *ptr, *arg_p, *opts;
 	substring_t args[MAX_OPT_ARGS];
 	int ret = 0, token;
+	unsigned long tmp_readonly;
 
 	opts = kstrdup(page, GFP_KERNEL);
 	if (!opts)
@@ -372,6 +453,22 @@ static ssize_t iblock_set_configfs_dev_params(struct se_hba *hba,
 					ib_dev->ibd_udev_path);
 			ib_dev->ibd_flags |= IBDF_HAS_UDEV_PATH;
 			break;
+		case Opt_readonly:
+			arg_p = match_strdup(&args[0]);
+			if (!arg_p) {
+				ret = -ENOMEM;
+				break;
+			}
+			ret = strict_strtoul(arg_p, 0, &tmp_readonly);
+			kfree(arg_p);
+			if (ret < 0) {
+				pr_err("strict_strtoul() failed for"
+						" readonly=\n");
+				goto out;
+			}
+			ib_dev->ibd_readonly = tmp_readonly;
+			pr_debug("IBLOCK: readonly: %d\n", ib_dev->ibd_readonly);
+			break;
 		case Opt_force:
 			break;
 		default:
@@ -411,11 +508,10 @@ static ssize_t iblock_show_configfs_dev_params(
 	if (bd)
 		bl += sprintf(b + bl, "iBlock device: %s",
 				bdevname(bd, buf));
-	if (ibd->ibd_flags & IBDF_HAS_UDEV_PATH) {
-		bl += sprintf(b + bl, "  UDEV PATH: %s\n",
+	if (ibd->ibd_flags & IBDF_HAS_UDEV_PATH)
+		bl += sprintf(b + bl, "  UDEV PATH: %s",
 				ibd->ibd_udev_path);
-	} else
-		bl += sprintf(b + bl, "\n");
+	bl += sprintf(b + bl, "  readonly: %d\n", ibd->ibd_readonly);
 
 	bl += sprintf(b + bl, "        ");
 	if (bd) {
@@ -493,9 +589,11 @@ static void iblock_submit_bios(struct bio_list *list, int rw)
 	blk_finish_plug(&plug);
 }
 
-static int iblock_execute_cmd(struct se_cmd *cmd, struct scatterlist *sgl,
-		u32 sgl_nents, enum dma_data_direction data_direction)
+static int iblock_execute_rw(struct se_cmd *cmd)
 {
+	struct scatterlist *sgl = cmd->t_data_sg;
+	u32 sgl_nents = cmd->t_data_nents;
+	enum dma_data_direction data_direction = cmd->data_direction;
 	struct se_device *dev = cmd->se_dev;
 	struct iblock_req *ibr;
 	struct bio *bio;
@@ -642,6 +740,18 @@ static void iblock_bio_done(struct bio *bio, int err)
 	iblock_complete_cmd(cmd);
 }
 
+static struct spc_ops iblock_spc_ops = {
+	.execute_rw		= iblock_execute_rw,
+	.execute_sync_cache	= iblock_execute_sync_cache,
+	.execute_write_same	= iblock_execute_write_same,
+	.execute_unmap		= iblock_execute_unmap,
+};
+
+static int iblock_parse_cdb(struct se_cmd *cmd)
+{
+	return sbc_parse_cdb(cmd, &iblock_spc_ops);
+}
+
 static struct se_subsystem_api iblock_template = {
 	.name			= "iblock",
 	.owner			= THIS_MODULE,
@@ -653,9 +763,7 @@ static struct se_subsystem_api iblock_template = {
 	.allocate_virtdevice	= iblock_allocate_virtdevice,
 	.create_virtdevice	= iblock_create_virtdevice,
 	.free_device		= iblock_free_device,
-	.execute_cmd		= iblock_execute_cmd,
-	.do_discard		= iblock_do_discard,
-	.do_sync_cache		= iblock_emulate_sync_cache,
+	.parse_cdb		= iblock_parse_cdb,
 	.check_configfs_dev_params = iblock_check_configfs_dev_params,
 	.set_configfs_dev_params = iblock_set_configfs_dev_params,
 	.show_configfs_dev_params = iblock_show_configfs_dev_params,
diff --git a/drivers/target/target_core_iblock.h b/drivers/target/target_core_iblock.h
index 66cf7b9e205e..533627ae79ec 100644
--- a/drivers/target/target_core_iblock.h
+++ b/drivers/target/target_core_iblock.h
@@ -18,6 +18,7 @@ struct iblock_dev {
 	u32	ibd_flags;
 	struct bio_set	*ibd_bio_set;
 	struct block_device *ibd_bd;
+	bool ibd_readonly;
 } ____cacheline_aligned;
 
 #endif /* TARGET_CORE_IBLOCK_H */
diff --git a/drivers/target/target_core_internal.h b/drivers/target/target_core_internal.h
index 165e82429687..0fd428225d11 100644
--- a/drivers/target/target_core_internal.h
+++ b/drivers/target/target_core_internal.h
@@ -4,25 +4,16 @@
 /* target_core_alua.c */
 extern struct t10_alua_lu_gp *default_lu_gp;
 
-/* target_core_cdb.c */
-int	target_emulate_inquiry(struct se_cmd *cmd);
-int	target_emulate_readcapacity(struct se_cmd *cmd);
-int	target_emulate_readcapacity_16(struct se_cmd *cmd);
-int	target_emulate_modesense(struct se_cmd *cmd);
-int	target_emulate_request_sense(struct se_cmd *cmd);
-int	target_emulate_unmap(struct se_cmd *cmd);
-int	target_emulate_write_same(struct se_cmd *cmd);
-int	target_emulate_synchronize_cache(struct se_cmd *cmd);
-int	target_emulate_noop(struct se_cmd *cmd);
-
 /* target_core_device.c */
 struct se_dev_entry *core_get_se_deve_from_rtpi(struct se_node_acl *, u16);
 int	core_free_device_list_for_node(struct se_node_acl *,
 		struct se_portal_group *);
 void	core_dec_lacl_count(struct se_node_acl *, struct se_cmd *);
 void	core_update_device_list_access(u32, u32, struct se_node_acl *);
-int	core_update_device_list_for_node(struct se_lun *, struct se_lun_acl *,
-		u32, u32, struct se_node_acl *, struct se_portal_group *, int);
+int	core_enable_device_list_for_node(struct se_lun *, struct se_lun_acl *,
+		u32, u32, struct se_node_acl *, struct se_portal_group *);
+int	core_disable_device_list_for_node(struct se_lun *, struct se_lun_acl *,
+		u32, u32, struct se_node_acl *, struct se_portal_group *);
 void	core_clear_lun_from_tpg(struct se_lun *, struct se_portal_group *);
 int	core_dev_export(struct se_device *, struct se_portal_group *,
 		struct se_lun *);
@@ -56,8 +47,7 @@ int	se_dev_set_max_sectors(struct se_device *, u32);
 int	se_dev_set_fabric_max_sectors(struct se_device *, u32);
 int	se_dev_set_optimal_sectors(struct se_device *, u32);
 int	se_dev_set_block_size(struct se_device *, u32);
-struct se_lun *core_dev_add_lun(struct se_portal_group *, struct se_hba *,
-		struct se_device *, u32);
+struct se_lun *core_dev_add_lun(struct se_portal_group *, struct se_device *, u32);
 int	core_dev_del_lun(struct se_portal_group *, u32);
 struct se_lun *core_get_lun_from_tpg(struct se_portal_group *, u32);
 struct se_lun_acl *core_dev_init_initiator_node_lun_acl(struct se_portal_group *,
@@ -104,7 +94,6 @@ void	release_se_kmem_caches(void);
 u32	scsi_get_new_index(scsi_index_t);
 void	transport_subsystem_check_init(void);
 void	transport_cmd_finish_abort(struct se_cmd *, int);
-void	__target_remove_from_execute_list(struct se_cmd *);
 unsigned char *transport_dump_cmd_direction(struct se_cmd *);
 void	transport_dump_dev_state(struct se_device *, char *, int *);
 void	transport_dump_dev_info(struct se_device *, struct se_lun *,
@@ -116,6 +105,7 @@ int	transport_dump_vpd_ident(struct t10_vpd *, unsigned char *, int);
 bool	target_stop_cmd(struct se_cmd *cmd, unsigned long *flags);
 int	transport_clear_lun_from_sessions(struct se_lun *);
 void	transport_send_task_abort(struct se_cmd *);
+int	target_cmd_size_check(struct se_cmd *cmd, unsigned int size);
 
 /* target_core_stat.c */
 void	target_stat_setup_dev_default_groups(struct se_subsystem_dev *);
diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c
index a1bcd927a9e6..1e946502c378 100644
--- a/drivers/target/target_core_pr.c
+++ b/drivers/target/target_core_pr.c
@@ -507,7 +507,7 @@ static int core_scsi3_pr_seq_non_holder(
 	 * Check if write exclusive initiator ports *NOT* holding the
 	 * WRITE_EXCLUSIVE_* reservation.
 	 */
-	if ((we) && !(registered_nexus)) {
+	if (we && !registered_nexus) {
 		if (cmd->data_direction == DMA_TO_DEVICE) {
 			/*
 			 * Conflict for write exclusive
@@ -2486,7 +2486,7 @@ static int core_scsi3_pro_reserve(
 	 */
 	spin_lock(&dev->dev_reservation_lock);
 	pr_res_holder = dev->dev_pr_res_holder;
-	if ((pr_res_holder)) {
+	if (pr_res_holder) {
 		/*
 		 * From spc4r17 Section 5.7.9: Reserving:
 		 *
@@ -4030,7 +4030,7 @@ static int core_scsi3_pri_read_reservation(struct se_cmd *cmd)
 
 	spin_lock(&se_dev->dev_reservation_lock);
 	pr_reg = se_dev->dev_pr_res_holder;
-	if ((pr_reg)) {
+	if (pr_reg) {
 		/*
 		 * Set the hardcoded Additional Length
 		 */
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index 4ce2cf642fce..6e32ff6f2fa0 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -35,8 +35,10 @@
 #include <linux/spinlock.h>
 #include <linux/genhd.h>
 #include <linux/cdrom.h>
-#include <linux/file.h>
+#include <linux/ratelimit.h>
 #include <linux/module.h>
+#include <asm/unaligned.h>
+
 #include <scsi/scsi.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_cmnd.h>
@@ -46,12 +48,14 @@
 #include <target/target_core_base.h>
 #include <target/target_core_backend.h>
 
+#include "target_core_alua.h"
 #include "target_core_pscsi.h"
 
 #define ISPRINT(a)  ((a >= ' ') && (a <= '~'))
 
 static struct se_subsystem_api pscsi_template;
 
+static int pscsi_execute_cmd(struct se_cmd *cmd);
 static void pscsi_req_done(struct request *, int);
 
 /*	pscsi_attach_hba():
@@ -1019,9 +1023,79 @@ fail:
 	return -ENOMEM;
 }
 
-static int pscsi_execute_cmd(struct se_cmd *cmd, struct scatterlist *sgl,
-		u32 sgl_nents, enum dma_data_direction data_direction)
+/*
+ * Clear a lun set in the cdb if the initiator talking to use spoke
+ * and old standards version, as we can't assume the underlying device
+ * won't choke up on it.
+ */
+static inline void pscsi_clear_cdb_lun(unsigned char *cdb)
+{
+	switch (cdb[0]) {
+	case READ_10: /* SBC - RDProtect */
+	case READ_12: /* SBC - RDProtect */
+	case READ_16: /* SBC - RDProtect */
+	case SEND_DIAGNOSTIC: /* SPC - SELF-TEST Code */
+	case VERIFY: /* SBC - VRProtect */
+	case VERIFY_16: /* SBC - VRProtect */
+	case WRITE_VERIFY: /* SBC - VRProtect */
+	case WRITE_VERIFY_12: /* SBC - VRProtect */
+	case MAINTENANCE_IN: /* SPC - Parameter Data Format for SA RTPG */
+		break;
+	default:
+		cdb[1] &= 0x1f; /* clear logical unit number */
+		break;
+	}
+}
+
+static int pscsi_parse_cdb(struct se_cmd *cmd)
+{
+	unsigned char *cdb = cmd->t_task_cdb;
+	unsigned int dummy_size;
+	int ret;
+
+	if (cmd->se_cmd_flags & SCF_BIDI) {
+		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		return -EINVAL;
+	}
+
+	pscsi_clear_cdb_lun(cdb);
+
+	/*
+	 * For REPORT LUNS we always need to emulate the response, for everything
+	 * else the default for pSCSI is to pass the command to the underlying
+	 * LLD / physical hardware.
+	 */
+	switch (cdb[0]) {
+	case REPORT_LUNS:
+		ret = spc_parse_cdb(cmd, &dummy_size);
+		if (ret)
+			return ret;
+		break;
+	case READ_6:
+	case READ_10:
+	case READ_12:
+	case READ_16:
+	case WRITE_6:
+	case WRITE_10:
+	case WRITE_12:
+	case WRITE_16:
+	case WRITE_VERIFY:
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		/* FALLTHROUGH*/
+	default:
+		cmd->execute_cmd = pscsi_execute_cmd;
+		break;
+	}
+
+	return 0;
+}
+
+static int pscsi_execute_cmd(struct se_cmd *cmd)
 {
+	struct scatterlist *sgl = cmd->t_data_sg;
+	u32 sgl_nents = cmd->t_data_nents;
+	enum dma_data_direction data_direction = cmd->data_direction;
 	struct pscsi_dev_virt *pdv = cmd->se_dev->dev_ptr;
 	struct pscsi_plugin_task *pt;
 	struct request *req;
@@ -1042,7 +1116,7 @@ static int pscsi_execute_cmd(struct se_cmd *cmd, struct scatterlist *sgl,
 	memcpy(pt->pscsi_cdb, cmd->t_task_cdb,
 		scsi_command_size(cmd->t_task_cdb));
 
-	if (cmd->se_cmd_flags & SCF_SCSI_NON_DATA_CDB) {
+	if (!sgl) {
 		req = blk_get_request(pdv->pdv_sd->request_queue,
 				(data_direction == DMA_TO_DEVICE),
 				GFP_KERNEL);
@@ -1188,7 +1262,7 @@ static struct se_subsystem_api pscsi_template = {
 	.create_virtdevice	= pscsi_create_virtdevice,
 	.free_device		= pscsi_free_device,
 	.transport_complete	= pscsi_transport_complete,
-	.execute_cmd		= pscsi_execute_cmd,
+	.parse_cdb		= pscsi_parse_cdb,
 	.check_configfs_dev_params = pscsi_check_configfs_dev_params,
 	.set_configfs_dev_params = pscsi_set_configfs_dev_params,
 	.show_configfs_dev_params = pscsi_show_configfs_dev_params,
diff --git a/drivers/target/target_core_rd.c b/drivers/target/target_core_rd.c
index d0ceb873c0e5..d00bbe33ff8b 100644
--- a/drivers/target/target_core_rd.c
+++ b/drivers/target/target_core_rd.c
@@ -284,9 +284,11 @@ static struct rd_dev_sg_table *rd_get_sg_table(struct rd_dev *rd_dev, u32 page)
 	return NULL;
 }
 
-static int rd_execute_cmd(struct se_cmd *cmd, struct scatterlist *sgl,
-		u32 sgl_nents, enum dma_data_direction data_direction)
+static int rd_execute_rw(struct se_cmd *cmd)
 {
+	struct scatterlist *sgl = cmd->t_data_sg;
+	u32 sgl_nents = cmd->t_data_nents;
+	enum dma_data_direction data_direction = cmd->data_direction;
 	struct se_device *se_dev = cmd->se_dev;
 	struct rd_dev *dev = se_dev->dev_ptr;
 	struct rd_dev_sg_table *table;
@@ -460,6 +462,15 @@ static sector_t rd_get_blocks(struct se_device *dev)
 	return blocks_long;
 }
 
+static struct spc_ops rd_spc_ops = {
+	.execute_rw		= rd_execute_rw,
+};
+
+static int rd_parse_cdb(struct se_cmd *cmd)
+{
+	return sbc_parse_cdb(cmd, &rd_spc_ops);
+}
+
 static struct se_subsystem_api rd_mcp_template = {
 	.name			= "rd_mcp",
 	.transport_type		= TRANSPORT_PLUGIN_VHBA_VDEV,
@@ -468,7 +479,7 @@ static struct se_subsystem_api rd_mcp_template = {
 	.allocate_virtdevice	= rd_allocate_virtdevice,
 	.create_virtdevice	= rd_create_virtdevice,
 	.free_device		= rd_free_device,
-	.execute_cmd		= rd_execute_cmd,
+	.parse_cdb		= rd_parse_cdb,
 	.check_configfs_dev_params = rd_check_configfs_dev_params,
 	.set_configfs_dev_params = rd_set_configfs_dev_params,
 	.show_configfs_dev_params = rd_show_configfs_dev_params,
diff --git a/drivers/target/target_core_sbc.c b/drivers/target/target_core_sbc.c
new file mode 100644
index 000000000000..a9dd9469e3bd
--- /dev/null
+++ b/drivers/target/target_core_sbc.c
@@ -0,0 +1,581 @@
+/*
+ * SCSI Block Commands (SBC) parsing and emulation.
+ *
+ * Copyright (c) 2002, 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/ratelimit.h>
+#include <asm/unaligned.h>
+#include <scsi/scsi.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_backend.h>
+#include <target/target_core_fabric.h>
+
+#include "target_core_internal.h"
+#include "target_core_ua.h"
+
+
+static int sbc_emulate_readcapacity(struct se_cmd *cmd)
+{
+	struct se_device *dev = cmd->se_dev;
+	unsigned char *buf;
+	unsigned long long blocks_long = dev->transport->get_blocks(dev);
+	u32 blocks;
+
+	if (blocks_long >= 0x00000000ffffffff)
+		blocks = 0xffffffff;
+	else
+		blocks = (u32)blocks_long;
+
+	buf = transport_kmap_data_sg(cmd);
+
+	buf[0] = (blocks >> 24) & 0xff;
+	buf[1] = (blocks >> 16) & 0xff;
+	buf[2] = (blocks >> 8) & 0xff;
+	buf[3] = blocks & 0xff;
+	buf[4] = (dev->se_sub_dev->se_dev_attrib.block_size >> 24) & 0xff;
+	buf[5] = (dev->se_sub_dev->se_dev_attrib.block_size >> 16) & 0xff;
+	buf[6] = (dev->se_sub_dev->se_dev_attrib.block_size >> 8) & 0xff;
+	buf[7] = dev->se_sub_dev->se_dev_attrib.block_size & 0xff;
+
+	transport_kunmap_data_sg(cmd);
+
+	target_complete_cmd(cmd, GOOD);
+	return 0;
+}
+
+static int sbc_emulate_readcapacity_16(struct se_cmd *cmd)
+{
+	struct se_device *dev = cmd->se_dev;
+	unsigned char *buf;
+	unsigned long long blocks = dev->transport->get_blocks(dev);
+
+	buf = transport_kmap_data_sg(cmd);
+
+	buf[0] = (blocks >> 56) & 0xff;
+	buf[1] = (blocks >> 48) & 0xff;
+	buf[2] = (blocks >> 40) & 0xff;
+	buf[3] = (blocks >> 32) & 0xff;
+	buf[4] = (blocks >> 24) & 0xff;
+	buf[5] = (blocks >> 16) & 0xff;
+	buf[6] = (blocks >> 8) & 0xff;
+	buf[7] = blocks & 0xff;
+	buf[8] = (dev->se_sub_dev->se_dev_attrib.block_size >> 24) & 0xff;
+	buf[9] = (dev->se_sub_dev->se_dev_attrib.block_size >> 16) & 0xff;
+	buf[10] = (dev->se_sub_dev->se_dev_attrib.block_size >> 8) & 0xff;
+	buf[11] = dev->se_sub_dev->se_dev_attrib.block_size & 0xff;
+	/*
+	 * Set Thin Provisioning Enable bit following sbc3r22 in section
+	 * READ CAPACITY (16) byte 14 if emulate_tpu or emulate_tpws is enabled.
+	 */
+	if (dev->se_sub_dev->se_dev_attrib.emulate_tpu || dev->se_sub_dev->se_dev_attrib.emulate_tpws)
+		buf[14] = 0x80;
+
+	transport_kunmap_data_sg(cmd);
+
+	target_complete_cmd(cmd, GOOD);
+	return 0;
+}
+
+int spc_get_write_same_sectors(struct se_cmd *cmd)
+{
+	u32 num_blocks;
+
+	if (cmd->t_task_cdb[0] == WRITE_SAME)
+		num_blocks = get_unaligned_be16(&cmd->t_task_cdb[7]);
+	else if (cmd->t_task_cdb[0] == WRITE_SAME_16)
+		num_blocks = get_unaligned_be32(&cmd->t_task_cdb[10]);
+	else /* WRITE_SAME_32 via VARIABLE_LENGTH_CMD */
+		num_blocks = get_unaligned_be32(&cmd->t_task_cdb[28]);
+
+	/*
+	 * Use the explicit range when non zero is supplied, otherwise calculate
+	 * the remaining range based on ->get_blocks() - starting LBA.
+	 */
+	if (num_blocks)
+		return num_blocks;
+
+	return cmd->se_dev->transport->get_blocks(cmd->se_dev) -
+		cmd->t_task_lba + 1;
+}
+EXPORT_SYMBOL(spc_get_write_same_sectors);
+
+static int sbc_emulate_verify(struct se_cmd *cmd)
+{
+	target_complete_cmd(cmd, GOOD);
+	return 0;
+}
+
+static inline u32 sbc_get_size(struct se_cmd *cmd, u32 sectors)
+{
+	return cmd->se_dev->se_sub_dev->se_dev_attrib.block_size * sectors;
+}
+
+static int sbc_check_valid_sectors(struct se_cmd *cmd)
+{
+	struct se_device *dev = cmd->se_dev;
+	unsigned long long end_lba;
+	u32 sectors;
+
+	sectors = cmd->data_length / dev->se_sub_dev->se_dev_attrib.block_size;
+	end_lba = dev->transport->get_blocks(dev) + 1;
+
+	if (cmd->t_task_lba + sectors > end_lba) {
+		pr_err("target: lba %llu, sectors %u exceeds end lba %llu\n",
+			cmd->t_task_lba, sectors, end_lba);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static inline u32 transport_get_sectors_6(unsigned char *cdb)
+{
+	/*
+	 * Use 8-bit sector value.  SBC-3 says:
+	 *
+	 *   A TRANSFER LENGTH field set to zero specifies that 256
+	 *   logical blocks shall be written.  Any other value
+	 *   specifies the number of logical blocks that shall be
+	 *   written.
+	 */
+	return cdb[4] ? : 256;
+}
+
+static inline u32 transport_get_sectors_10(unsigned char *cdb)
+{
+	return (u32)(cdb[7] << 8) + cdb[8];
+}
+
+static inline u32 transport_get_sectors_12(unsigned char *cdb)
+{
+	return (u32)(cdb[6] << 24) + (cdb[7] << 16) + (cdb[8] << 8) + cdb[9];
+}
+
+static inline u32 transport_get_sectors_16(unsigned char *cdb)
+{
+	return (u32)(cdb[10] << 24) + (cdb[11] << 16) +
+		    (cdb[12] << 8) + cdb[13];
+}
+
+/*
+ * Used for VARIABLE_LENGTH_CDB WRITE_32 and READ_32 variants
+ */
+static inline u32 transport_get_sectors_32(unsigned char *cdb)
+{
+	return (u32)(cdb[28] << 24) + (cdb[29] << 16) +
+		    (cdb[30] << 8) + cdb[31];
+
+}
+
+static inline u32 transport_lba_21(unsigned char *cdb)
+{
+	return ((cdb[1] & 0x1f) << 16) | (cdb[2] << 8) | cdb[3];
+}
+
+static inline u32 transport_lba_32(unsigned char *cdb)
+{
+	return (cdb[2] << 24) | (cdb[3] << 16) | (cdb[4] << 8) | cdb[5];
+}
+
+static inline unsigned long long transport_lba_64(unsigned char *cdb)
+{
+	unsigned int __v1, __v2;
+
+	__v1 = (cdb[2] << 24) | (cdb[3] << 16) | (cdb[4] << 8) | cdb[5];
+	__v2 = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
+
+	return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
+}
+
+/*
+ * For VARIABLE_LENGTH_CDB w/ 32 byte extended CDBs
+ */
+static inline unsigned long long transport_lba_64_ext(unsigned char *cdb)
+{
+	unsigned int __v1, __v2;
+
+	__v1 = (cdb[12] << 24) | (cdb[13] << 16) | (cdb[14] << 8) | cdb[15];
+	__v2 = (cdb[16] << 24) | (cdb[17] << 16) | (cdb[18] << 8) | cdb[19];
+
+	return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
+}
+
+static int sbc_write_same_supported(struct se_device *dev,
+		unsigned char *flags)
+{
+	if ((flags[0] & 0x04) || (flags[0] & 0x02)) {
+		pr_err("WRITE_SAME PBDATA and LBDATA"
+			" bits not supported for Block Discard"
+			" Emulation\n");
+		return -ENOSYS;
+	}
+
+	/*
+	 * Currently for the emulated case we only accept
+	 * tpws with the UNMAP=1 bit set.
+	 */
+	if (!(flags[0] & 0x08)) {
+		pr_err("WRITE_SAME w/o UNMAP bit not"
+			" supported for Block Discard Emulation\n");
+		return -ENOSYS;
+	}
+
+	return 0;
+}
+
+static void xdreadwrite_callback(struct se_cmd *cmd)
+{
+	unsigned char *buf, *addr;
+	struct scatterlist *sg;
+	unsigned int offset;
+	int i;
+	int count;
+	/*
+	 * From sbc3r22.pdf section 5.48 XDWRITEREAD (10) command
+	 *
+	 * 1) read the specified logical block(s);
+	 * 2) transfer logical blocks from the data-out buffer;
+	 * 3) XOR the logical blocks transferred from the data-out buffer with
+	 *    the logical blocks read, storing the resulting XOR data in a buffer;
+	 * 4) if the DISABLE WRITE bit is set to zero, then write the logical
+	 *    blocks transferred from the data-out buffer; and
+	 * 5) transfer the resulting XOR data to the data-in buffer.
+	 */
+	buf = kmalloc(cmd->data_length, GFP_KERNEL);
+	if (!buf) {
+		pr_err("Unable to allocate xor_callback buf\n");
+		return;
+	}
+	/*
+	 * Copy the scatterlist WRITE buffer located at cmd->t_data_sg
+	 * into the locally allocated *buf
+	 */
+	sg_copy_to_buffer(cmd->t_data_sg,
+			  cmd->t_data_nents,
+			  buf,
+			  cmd->data_length);
+
+	/*
+	 * Now perform the XOR against the BIDI read memory located at
+	 * cmd->t_mem_bidi_list
+	 */
+
+	offset = 0;
+	for_each_sg(cmd->t_bidi_data_sg, sg, cmd->t_bidi_data_nents, count) {
+		addr = kmap_atomic(sg_page(sg));
+		if (!addr)
+			goto out;
+
+		for (i = 0; i < sg->length; i++)
+			*(addr + sg->offset + i) ^= *(buf + offset + i);
+
+		offset += sg->length;
+		kunmap_atomic(addr);
+	}
+
+out:
+	kfree(buf);
+}
+
+int sbc_parse_cdb(struct se_cmd *cmd, struct spc_ops *ops)
+{
+	struct se_subsystem_dev *su_dev = cmd->se_dev->se_sub_dev;
+	struct se_device *dev = cmd->se_dev;
+	unsigned char *cdb = cmd->t_task_cdb;
+	unsigned int size;
+	u32 sectors = 0;
+	int ret;
+
+	switch (cdb[0]) {
+	case READ_6:
+		sectors = transport_get_sectors_6(cdb);
+		cmd->t_task_lba = transport_lba_21(cdb);
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case READ_10:
+		sectors = transport_get_sectors_10(cdb);
+		cmd->t_task_lba = transport_lba_32(cdb);
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case READ_12:
+		sectors = transport_get_sectors_12(cdb);
+		cmd->t_task_lba = transport_lba_32(cdb);
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case READ_16:
+		sectors = transport_get_sectors_16(cdb);
+		cmd->t_task_lba = transport_lba_64(cdb);
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case WRITE_6:
+		sectors = transport_get_sectors_6(cdb);
+		cmd->t_task_lba = transport_lba_21(cdb);
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case WRITE_10:
+	case WRITE_VERIFY:
+		sectors = transport_get_sectors_10(cdb);
+		cmd->t_task_lba = transport_lba_32(cdb);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case WRITE_12:
+		sectors = transport_get_sectors_12(cdb);
+		cmd->t_task_lba = transport_lba_32(cdb);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case WRITE_16:
+		sectors = transport_get_sectors_16(cdb);
+		cmd->t_task_lba = transport_lba_64(cdb);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+		cmd->execute_cmd = ops->execute_rw;
+		break;
+	case XDWRITEREAD_10:
+		if ((cmd->data_direction != DMA_TO_DEVICE) ||
+		    !(cmd->se_cmd_flags & SCF_BIDI))
+			goto out_invalid_cdb_field;
+		sectors = transport_get_sectors_10(cdb);
+
+		cmd->t_task_lba = transport_lba_32(cdb);
+		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+
+		/*
+		 * Setup BIDI XOR callback to be run after I/O completion.
+		 */
+		cmd->execute_cmd = ops->execute_rw;
+		cmd->transport_complete_callback = &xdreadwrite_callback;
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
+		break;
+	case VARIABLE_LENGTH_CMD:
+	{
+		u16 service_action = get_unaligned_be16(&cdb[8]);
+		switch (service_action) {
+		case XDWRITEREAD_32:
+			sectors = transport_get_sectors_32(cdb);
+
+			/*
+			 * Use WRITE_32 and READ_32 opcodes for the emulated
+			 * XDWRITE_READ_32 logic.
+			 */
+			cmd->t_task_lba = transport_lba_64_ext(cdb);
+			cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
+
+			/*
+			 * Setup BIDI XOR callback to be run during after I/O
+			 * completion.
+			 */
+			cmd->execute_cmd = ops->execute_rw;
+			cmd->transport_complete_callback = &xdreadwrite_callback;
+			if (cdb[1] & 0x8)
+				cmd->se_cmd_flags |= SCF_FUA;
+			break;
+		case WRITE_SAME_32:
+			if (!ops->execute_write_same)
+				goto out_unsupported_cdb;
+
+			sectors = transport_get_sectors_32(cdb);
+			if (!sectors) {
+				pr_err("WSNZ=1, WRITE_SAME w/sectors=0 not"
+				       " supported\n");
+				goto out_invalid_cdb_field;
+			}
+
+			size = sbc_get_size(cmd, 1);
+			cmd->t_task_lba = get_unaligned_be64(&cdb[12]);
+
+			if (sbc_write_same_supported(dev, &cdb[10]) < 0)
+				goto out_unsupported_cdb;
+			cmd->execute_cmd = ops->execute_write_same;
+			break;
+		default:
+			pr_err("VARIABLE_LENGTH_CMD service action"
+				" 0x%04x not supported\n", service_action);
+			goto out_unsupported_cdb;
+		}
+		break;
+	}
+	case READ_CAPACITY:
+		size = READ_CAP_LEN;
+		cmd->execute_cmd = sbc_emulate_readcapacity;
+		break;
+	case SERVICE_ACTION_IN:
+		switch (cmd->t_task_cdb[1] & 0x1f) {
+		case SAI_READ_CAPACITY_16:
+			cmd->execute_cmd = sbc_emulate_readcapacity_16;
+			break;
+		default:
+			pr_err("Unsupported SA: 0x%02x\n",
+				cmd->t_task_cdb[1] & 0x1f);
+			goto out_invalid_cdb_field;
+		}
+		size = (cdb[10] << 24) | (cdb[11] << 16) |
+		       (cdb[12] << 8) | cdb[13];
+		break;
+	case SYNCHRONIZE_CACHE:
+	case SYNCHRONIZE_CACHE_16:
+		if (!ops->execute_sync_cache)
+			goto out_unsupported_cdb;
+
+		/*
+		 * Extract LBA and range to be flushed for emulated SYNCHRONIZE_CACHE
+		 */
+		if (cdb[0] == SYNCHRONIZE_CACHE) {
+			sectors = transport_get_sectors_10(cdb);
+			cmd->t_task_lba = transport_lba_32(cdb);
+		} else {
+			sectors = transport_get_sectors_16(cdb);
+			cmd->t_task_lba = transport_lba_64(cdb);
+		}
+
+		size = sbc_get_size(cmd, sectors);
+
+		/*
+		 * Check to ensure that LBA + Range does not exceed past end of
+		 * device for IBLOCK and FILEIO ->do_sync_cache() backend calls
+		 */
+		if (cmd->t_task_lba || sectors) {
+			if (sbc_check_valid_sectors(cmd) < 0)
+				goto out_invalid_cdb_field;
+		}
+		cmd->execute_cmd = ops->execute_sync_cache;
+		break;
+	case UNMAP:
+		if (!ops->execute_unmap)
+			goto out_unsupported_cdb;
+
+		size = get_unaligned_be16(&cdb[7]);
+		cmd->execute_cmd = ops->execute_unmap;
+		break;
+	case WRITE_SAME_16:
+		if (!ops->execute_write_same)
+			goto out_unsupported_cdb;
+
+		sectors = transport_get_sectors_16(cdb);
+		if (!sectors) {
+			pr_err("WSNZ=1, WRITE_SAME w/sectors=0 not supported\n");
+			goto out_invalid_cdb_field;
+		}
+
+		size = sbc_get_size(cmd, 1);
+		cmd->t_task_lba = get_unaligned_be64(&cdb[2]);
+
+		if (sbc_write_same_supported(dev, &cdb[1]) < 0)
+			goto out_unsupported_cdb;
+		cmd->execute_cmd = ops->execute_write_same;
+		break;
+	case WRITE_SAME:
+		if (!ops->execute_write_same)
+			goto out_unsupported_cdb;
+
+		sectors = transport_get_sectors_10(cdb);
+		if (!sectors) {
+			pr_err("WSNZ=1, WRITE_SAME w/sectors=0 not supported\n");
+			goto out_invalid_cdb_field;
+		}
+
+		size = sbc_get_size(cmd, 1);
+		cmd->t_task_lba = get_unaligned_be32(&cdb[2]);
+
+		/*
+		 * Follow sbcr26 with WRITE_SAME (10) and check for the existence
+		 * of byte 1 bit 3 UNMAP instead of original reserved field
+		 */
+		if (sbc_write_same_supported(dev, &cdb[1]) < 0)
+			goto out_unsupported_cdb;
+		cmd->execute_cmd = ops->execute_write_same;
+		break;
+	case VERIFY:
+		size = 0;
+		cmd->execute_cmd = sbc_emulate_verify;
+		break;
+	default:
+		ret = spc_parse_cdb(cmd, &size);
+		if (ret)
+			return ret;
+	}
+
+	/* reject any command that we don't have a handler for */
+	if (!(cmd->se_cmd_flags & SCF_SCSI_DATA_CDB) && !cmd->execute_cmd)
+		goto out_unsupported_cdb;
+
+	if (cmd->se_cmd_flags & SCF_SCSI_DATA_CDB) {
+		unsigned long long end_lba;
+
+		if (sectors > su_dev->se_dev_attrib.fabric_max_sectors) {
+			printk_ratelimited(KERN_ERR "SCSI OP %02xh with too"
+				" big sectors %u exceeds fabric_max_sectors:"
+				" %u\n", cdb[0], sectors,
+				su_dev->se_dev_attrib.fabric_max_sectors);
+			goto out_invalid_cdb_field;
+		}
+		if (sectors > su_dev->se_dev_attrib.hw_max_sectors) {
+			printk_ratelimited(KERN_ERR "SCSI OP %02xh with too"
+				" big sectors %u exceeds backend hw_max_sectors:"
+				" %u\n", cdb[0], sectors,
+				su_dev->se_dev_attrib.hw_max_sectors);
+			goto out_invalid_cdb_field;
+		}
+
+		end_lba = dev->transport->get_blocks(dev) + 1;
+		if (cmd->t_task_lba + sectors > end_lba) {
+			pr_err("cmd exceeds last lba %llu "
+				"(lba %llu, sectors %u)\n",
+				end_lba, cmd->t_task_lba, sectors);
+			goto out_invalid_cdb_field;
+		}
+
+		size = sbc_get_size(cmd, sectors);
+	}
+
+	ret = target_cmd_size_check(cmd, size);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+
+out_unsupported_cdb:
+	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+	cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+	return -EINVAL;
+out_invalid_cdb_field:
+	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+	cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+	return -EINVAL;
+}
+EXPORT_SYMBOL(sbc_parse_cdb);
diff --git a/drivers/target/target_core_cdb.c b/drivers/target/target_core_spc.c
index 664f6e775d0e..4c861de538c9 100644
--- a/drivers/target/target_core_cdb.c
+++ b/drivers/target/target_core_spc.c
@@ -1,5 +1,5 @@
 /*
- * CDB emulation for non-READ/WRITE commands.
+ * SCSI Primary Commands (SPC) parsing and emulation.
  *
  * Copyright (c) 2002, 2003, 2004, 2005 PyX Technologies, Inc.
  * Copyright (c) 2005, 2006, 2007 SBE, Inc.
@@ -26,17 +26,21 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <asm/unaligned.h>
+
 #include <scsi/scsi.h>
+#include <scsi/scsi_tcq.h>
 
 #include <target/target_core_base.h>
 #include <target/target_core_backend.h>
 #include <target/target_core_fabric.h>
 
 #include "target_core_internal.h"
+#include "target_core_alua.h"
+#include "target_core_pr.h"
 #include "target_core_ua.h"
 
-static void
-target_fill_alua_data(struct se_port *port, unsigned char *buf)
+
+static void spc_fill_alua_data(struct se_port *port, unsigned char *buf)
 {
 	struct t10_alua_tg_pt_gp *tg_pt_gp;
 	struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
@@ -65,8 +69,7 @@ target_fill_alua_data(struct se_port *port, unsigned char *buf)
 	spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
 }
 
-static int
-target_emulate_inquiry_std(struct se_cmd *cmd, char *buf)
+static int spc_emulate_inquiry_std(struct se_cmd *cmd, char *buf)
 {
 	struct se_lun *lun = cmd->se_lun;
 	struct se_device *dev = cmd->se_dev;
@@ -93,7 +96,7 @@ target_emulate_inquiry_std(struct se_cmd *cmd, char *buf)
 	 * Enable SCCS and TPGS fields for Emulated ALUA
 	 */
 	if (dev->se_sub_dev->t10_alua.alua_type == SPC3_ALUA_EMULATED)
-		target_fill_alua_data(lun->lun_sep, buf);
+		spc_fill_alua_data(lun->lun_sep, buf);
 
 	buf[7] = 0x2; /* CmdQue=1 */
 
@@ -106,8 +109,7 @@ target_emulate_inquiry_std(struct se_cmd *cmd, char *buf)
 }
 
 /* unit serial number */
-static int
-target_emulate_evpd_80(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_80(struct se_cmd *cmd, unsigned char *buf)
 {
 	struct se_device *dev = cmd->se_dev;
 	u16 len = 0;
@@ -127,8 +129,8 @@ target_emulate_evpd_80(struct se_cmd *cmd, unsigned char *buf)
 	return 0;
 }
 
-static void
-target_parse_naa_6h_vendor_specific(struct se_device *dev, unsigned char *buf)
+static void spc_parse_naa_6h_vendor_specific(struct se_device *dev,
+		unsigned char *buf)
 {
 	unsigned char *p = &dev->se_sub_dev->t10_wwn.unit_serial[0];
 	int cnt;
@@ -162,8 +164,7 @@ target_parse_naa_6h_vendor_specific(struct se_device *dev, unsigned char *buf)
  * Device identification VPD, for a complete list of
  * DESIGNATOR TYPEs see spc4r17 Table 459.
  */
-static int
-target_emulate_evpd_83(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_83(struct se_cmd *cmd, unsigned char *buf)
 {
 	struct se_device *dev = cmd->se_dev;
 	struct se_lun *lun = cmd->se_lun;
@@ -220,7 +221,7 @@ target_emulate_evpd_83(struct se_cmd *cmd, unsigned char *buf)
 	 * VENDOR_SPECIFIC_IDENTIFIER and
 	 * VENDOR_SPECIFIC_IDENTIFIER_EXTENTION
 	 */
-	target_parse_naa_6h_vendor_specific(dev, &buf[off]);
+	spc_parse_naa_6h_vendor_specific(dev, &buf[off]);
 
 	len = 20;
 	off = (len + 4);
@@ -414,8 +415,7 @@ check_scsi_name:
 }
 
 /* Extended INQUIRY Data VPD Page */
-static int
-target_emulate_evpd_86(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_86(struct se_cmd *cmd, unsigned char *buf)
 {
 	buf[3] = 0x3c;
 	/* Set HEADSUP, ORDSUP, SIMPSUP */
@@ -428,15 +428,14 @@ target_emulate_evpd_86(struct se_cmd *cmd, unsigned char *buf)
 }
 
 /* Block Limits VPD page */
-static int
-target_emulate_evpd_b0(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_b0(struct se_cmd *cmd, unsigned char *buf)
 {
 	struct se_device *dev = cmd->se_dev;
 	u32 max_sectors;
 	int have_tp = 0;
 
 	/*
-	 * Following sbc3r22 section 6.5.3 Block Limits VPD page, when
+	 * Following spc3r22 section 6.5.3 Block Limits VPD page, when
 	 * emulate_tpu=1 or emulate_tpws=1 we will be expect a
 	 * different page length for Thin Provisioning.
 	 */
@@ -500,8 +499,7 @@ target_emulate_evpd_b0(struct se_cmd *cmd, unsigned char *buf)
 }
 
 /* Block Device Characteristics VPD page */
-static int
-target_emulate_evpd_b1(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_b1(struct se_cmd *cmd, unsigned char *buf)
 {
 	struct se_device *dev = cmd->se_dev;
 
@@ -513,13 +511,12 @@ target_emulate_evpd_b1(struct se_cmd *cmd, unsigned char *buf)
 }
 
 /* Thin Provisioning VPD */
-static int
-target_emulate_evpd_b2(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_b2(struct se_cmd *cmd, unsigned char *buf)
 {
 	struct se_device *dev = cmd->se_dev;
 
 	/*
-	 * From sbc3r22 section 6.5.4 Thin Provisioning VPD page:
+	 * From spc3r22 section 6.5.4 Thin Provisioning VPD page:
 	 *
 	 * The PAGE LENGTH field is defined in SPC-4. If the DP bit is set to
 	 * zero, then the page length shall be set to 0004h.  If the DP bit
@@ -564,25 +561,23 @@ target_emulate_evpd_b2(struct se_cmd *cmd, unsigned char *buf)
 	return 0;
 }
 
-static int
-target_emulate_evpd_00(struct se_cmd *cmd, unsigned char *buf);
+static int spc_emulate_evpd_00(struct se_cmd *cmd, unsigned char *buf);
 
 static struct {
 	uint8_t		page;
 	int		(*emulate)(struct se_cmd *, unsigned char *);
 } evpd_handlers[] = {
-	{ .page = 0x00, .emulate = target_emulate_evpd_00 },
-	{ .page = 0x80, .emulate = target_emulate_evpd_80 },
-	{ .page = 0x83, .emulate = target_emulate_evpd_83 },
-	{ .page = 0x86, .emulate = target_emulate_evpd_86 },
-	{ .page = 0xb0, .emulate = target_emulate_evpd_b0 },
-	{ .page = 0xb1, .emulate = target_emulate_evpd_b1 },
-	{ .page = 0xb2, .emulate = target_emulate_evpd_b2 },
+	{ .page = 0x00, .emulate = spc_emulate_evpd_00 },
+	{ .page = 0x80, .emulate = spc_emulate_evpd_80 },
+	{ .page = 0x83, .emulate = spc_emulate_evpd_83 },
+	{ .page = 0x86, .emulate = spc_emulate_evpd_86 },
+	{ .page = 0xb0, .emulate = spc_emulate_evpd_b0 },
+	{ .page = 0xb1, .emulate = spc_emulate_evpd_b1 },
+	{ .page = 0xb2, .emulate = spc_emulate_evpd_b2 },
 };
 
 /* supported vital product data pages */
-static int
-target_emulate_evpd_00(struct se_cmd *cmd, unsigned char *buf)
+static int spc_emulate_evpd_00(struct se_cmd *cmd, unsigned char *buf)
 {
 	int p;
 
@@ -601,7 +596,7 @@ target_emulate_evpd_00(struct se_cmd *cmd, unsigned char *buf)
 	return 0;
 }
 
-int target_emulate_inquiry(struct se_cmd *cmd)
+static int spc_emulate_inquiry(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
 	struct se_portal_group *tpg = cmd->se_lun->lun_sep->sep_tpg;
@@ -643,7 +638,7 @@ int target_emulate_inquiry(struct se_cmd *cmd)
 			goto out;
 		}
 
-		ret = target_emulate_inquiry_std(cmd, buf);
+		ret = spc_emulate_inquiry_std(cmd, buf);
 		goto out;
 	}
 
@@ -671,70 +666,7 @@ out:
 	return ret;
 }
 
-int target_emulate_readcapacity(struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-	unsigned char *buf;
-	unsigned long long blocks_long = dev->transport->get_blocks(dev);
-	u32 blocks;
-
-	if (blocks_long >= 0x00000000ffffffff)
-		blocks = 0xffffffff;
-	else
-		blocks = (u32)blocks_long;
-
-	buf = transport_kmap_data_sg(cmd);
-
-	buf[0] = (blocks >> 24) & 0xff;
-	buf[1] = (blocks >> 16) & 0xff;
-	buf[2] = (blocks >> 8) & 0xff;
-	buf[3] = blocks & 0xff;
-	buf[4] = (dev->se_sub_dev->se_dev_attrib.block_size >> 24) & 0xff;
-	buf[5] = (dev->se_sub_dev->se_dev_attrib.block_size >> 16) & 0xff;
-	buf[6] = (dev->se_sub_dev->se_dev_attrib.block_size >> 8) & 0xff;
-	buf[7] = dev->se_sub_dev->se_dev_attrib.block_size & 0xff;
-
-	transport_kunmap_data_sg(cmd);
-
-	target_complete_cmd(cmd, GOOD);
-	return 0;
-}
-
-int target_emulate_readcapacity_16(struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-	unsigned char *buf;
-	unsigned long long blocks = dev->transport->get_blocks(dev);
-
-	buf = transport_kmap_data_sg(cmd);
-
-	buf[0] = (blocks >> 56) & 0xff;
-	buf[1] = (blocks >> 48) & 0xff;
-	buf[2] = (blocks >> 40) & 0xff;
-	buf[3] = (blocks >> 32) & 0xff;
-	buf[4] = (blocks >> 24) & 0xff;
-	buf[5] = (blocks >> 16) & 0xff;
-	buf[6] = (blocks >> 8) & 0xff;
-	buf[7] = blocks & 0xff;
-	buf[8] = (dev->se_sub_dev->se_dev_attrib.block_size >> 24) & 0xff;
-	buf[9] = (dev->se_sub_dev->se_dev_attrib.block_size >> 16) & 0xff;
-	buf[10] = (dev->se_sub_dev->se_dev_attrib.block_size >> 8) & 0xff;
-	buf[11] = dev->se_sub_dev->se_dev_attrib.block_size & 0xff;
-	/*
-	 * Set Thin Provisioning Enable bit following sbc3r22 in section
-	 * READ CAPACITY (16) byte 14 if emulate_tpu or emulate_tpws is enabled.
-	 */
-	if (dev->se_sub_dev->se_dev_attrib.emulate_tpu || dev->se_sub_dev->se_dev_attrib.emulate_tpws)
-		buf[14] = 0x80;
-
-	transport_kunmap_data_sg(cmd);
-
-	target_complete_cmd(cmd, GOOD);
-	return 0;
-}
-
-static int
-target_modesense_rwrecovery(unsigned char *p)
+static int spc_modesense_rwrecovery(unsigned char *p)
 {
 	p[0] = 0x01;
 	p[1] = 0x0a;
@@ -742,8 +674,7 @@ target_modesense_rwrecovery(unsigned char *p)
 	return 12;
 }
 
-static int
-target_modesense_control(struct se_device *dev, unsigned char *p)
+static int spc_modesense_control(struct se_device *dev, unsigned char *p)
 {
 	p[0] = 0x0a;
 	p[1] = 0x0a;
@@ -828,8 +759,7 @@ target_modesense_control(struct se_device *dev, unsigned char *p)
 	return 12;
 }
 
-static int
-target_modesense_caching(struct se_device *dev, unsigned char *p)
+static int spc_modesense_caching(struct se_device *dev, unsigned char *p)
 {
 	p[0] = 0x08;
 	p[1] = 0x12;
@@ -840,8 +770,7 @@ target_modesense_caching(struct se_device *dev, unsigned char *p)
 	return 20;
 }
 
-static void
-target_modesense_write_protect(unsigned char *buf, int type)
+static void spc_modesense_write_protect(unsigned char *buf, int type)
 {
 	/*
 	 * I believe that the WP bit (bit 7) in the mode header is the same for
@@ -856,8 +785,7 @@ target_modesense_write_protect(unsigned char *buf, int type)
 	}
 }
 
-static void
-target_modesense_dpofua(unsigned char *buf, int type)
+static void spc_modesense_dpofua(unsigned char *buf, int type)
 {
 	switch (type) {
 	case TYPE_DISK:
@@ -868,7 +796,7 @@ target_modesense_dpofua(unsigned char *buf, int type)
 	}
 }
 
-int target_emulate_modesense(struct se_cmd *cmd)
+static int spc_emulate_modesense(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
 	char *cdb = cmd->t_task_cdb;
@@ -883,18 +811,18 @@ int target_emulate_modesense(struct se_cmd *cmd)
 
 	switch (cdb[2] & 0x3f) {
 	case 0x01:
-		length = target_modesense_rwrecovery(&buf[offset]);
+		length = spc_modesense_rwrecovery(&buf[offset]);
 		break;
 	case 0x08:
-		length = target_modesense_caching(dev, &buf[offset]);
+		length = spc_modesense_caching(dev, &buf[offset]);
 		break;
 	case 0x0a:
-		length = target_modesense_control(dev, &buf[offset]);
+		length = spc_modesense_control(dev, &buf[offset]);
 		break;
 	case 0x3f:
-		length = target_modesense_rwrecovery(&buf[offset]);
-		length += target_modesense_caching(dev, &buf[offset+length]);
-		length += target_modesense_control(dev, &buf[offset+length]);
+		length = spc_modesense_rwrecovery(&buf[offset]);
+		length += spc_modesense_caching(dev, &buf[offset+length]);
+		length += spc_modesense_control(dev, &buf[offset+length]);
 		break;
 	default:
 		pr_err("MODE SENSE: unimplemented page/subpage: 0x%02x/0x%02x\n",
@@ -912,11 +840,11 @@ int target_emulate_modesense(struct se_cmd *cmd)
 		if ((cmd->se_lun->lun_access & TRANSPORT_LUNFLAGS_READ_ONLY) ||
 		    (cmd->se_deve &&
 		    (cmd->se_deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY)))
-			target_modesense_write_protect(&buf[3], type);
+			spc_modesense_write_protect(&buf[3], type);
 
 		if ((dev->se_sub_dev->se_dev_attrib.emulate_write_cache > 0) &&
 		    (dev->se_sub_dev->se_dev_attrib.emulate_fua_write > 0))
-			target_modesense_dpofua(&buf[3], type);
+			spc_modesense_dpofua(&buf[3], type);
 
 		if ((offset + 2) > cmd->data_length)
 			offset = cmd->data_length;
@@ -928,11 +856,11 @@ int target_emulate_modesense(struct se_cmd *cmd)
 		if ((cmd->se_lun->lun_access & TRANSPORT_LUNFLAGS_READ_ONLY) ||
 		    (cmd->se_deve &&
 		    (cmd->se_deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY)))
-			target_modesense_write_protect(&buf[2], type);
+			spc_modesense_write_protect(&buf[2], type);
 
 		if ((dev->se_sub_dev->se_dev_attrib.emulate_write_cache > 0) &&
 		    (dev->se_sub_dev->se_dev_attrib.emulate_fua_write > 0))
-			target_modesense_dpofua(&buf[2], type);
+			spc_modesense_dpofua(&buf[2], type);
 
 		if ((offset + 1) > cmd->data_length)
 			offset = cmd->data_length;
@@ -946,7 +874,7 @@ int target_emulate_modesense(struct se_cmd *cmd)
 	return 0;
 }
 
-int target_emulate_request_sense(struct se_cmd *cmd)
+static int spc_emulate_request_sense(struct se_cmd *cmd)
 {
 	unsigned char *cdb = cmd->t_task_cdb;
 	unsigned char *buf;
@@ -1005,126 +933,172 @@ end:
 	return 0;
 }
 
-/*
- * Used for TCM/IBLOCK and TCM/FILEIO for block/blk-lib.c level discard support.
- * Note this is not used for TCM/pSCSI passthrough
- */
-int target_emulate_unmap(struct se_cmd *cmd)
+static int spc_emulate_testunitready(struct se_cmd *cmd)
 {
-	struct se_device *dev = cmd->se_dev;
-	unsigned char *buf, *ptr = NULL;
-	unsigned char *cdb = &cmd->t_task_cdb[0];
-	sector_t lba;
-	unsigned int size = cmd->data_length, range;
-	int ret = 0, offset;
-	unsigned short dl, bd_dl;
-
-	if (!dev->transport->do_discard) {
-		pr_err("UNMAP emulation not supported for: %s\n",
-				dev->transport->name);
-		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
-		return -ENOSYS;
-	}
-
-	/* First UNMAP block descriptor starts at 8 byte offset */
-	offset = 8;
-	size -= 8;
-	dl = get_unaligned_be16(&cdb[0]);
-	bd_dl = get_unaligned_be16(&cdb[2]);
-
-	buf = transport_kmap_data_sg(cmd);
-
-	ptr = &buf[offset];
-	pr_debug("UNMAP: Sub: %s Using dl: %hu bd_dl: %hu size: %hu"
-		" ptr: %p\n", dev->transport->name, dl, bd_dl, size, ptr);
-
-	while (size) {
-		lba = get_unaligned_be64(&ptr[0]);
-		range = get_unaligned_be32(&ptr[8]);
-		pr_debug("UNMAP: Using lba: %llu and range: %u\n",
-				 (unsigned long long)lba, range);
-
-		ret = dev->transport->do_discard(dev, lba, range);
-		if (ret < 0) {
-			pr_err("blkdev_issue_discard() failed: %d\n",
-					ret);
-			goto err;
-		}
-
-		ptr += 16;
-		size -= 16;
-	}
-
-err:
-	transport_kunmap_data_sg(cmd);
-	if (!ret)
-		target_complete_cmd(cmd, GOOD);
-	return ret;
+	target_complete_cmd(cmd, GOOD);
+	return 0;
 }
 
-/*
- * Used for TCM/IBLOCK and TCM/FILEIO for block/blk-lib.c level discard support.
- * Note this is not used for TCM/pSCSI passthrough
- */
-int target_emulate_write_same(struct se_cmd *cmd)
+int spc_parse_cdb(struct se_cmd *cmd, unsigned int *size)
 {
 	struct se_device *dev = cmd->se_dev;
-	sector_t range;
-	sector_t lba = cmd->t_task_lba;
-	u32 num_blocks;
-	int ret;
-
-	if (!dev->transport->do_discard) {
-		pr_err("WRITE_SAME emulation not supported"
-				" for: %s\n", dev->transport->name);
-		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
-		return -ENOSYS;
-	}
-
-	if (cmd->t_task_cdb[0] == WRITE_SAME)
-		num_blocks = get_unaligned_be16(&cmd->t_task_cdb[7]);
-	else if (cmd->t_task_cdb[0] == WRITE_SAME_16)
-		num_blocks = get_unaligned_be32(&cmd->t_task_cdb[10]);
-	else /* WRITE_SAME_32 via VARIABLE_LENGTH_CMD */
-		num_blocks = get_unaligned_be32(&cmd->t_task_cdb[28]);
-
-	/*
-	 * Use the explicit range when non zero is supplied, otherwise calculate
-	 * the remaining range based on ->get_blocks() - starting LBA.
-	 */
-	if (num_blocks != 0)
-		range = num_blocks;
-	else
-		range = (dev->transport->get_blocks(dev) - lba) + 1;
-
-	pr_debug("WRITE_SAME UNMAP: LBA: %llu Range: %llu\n",
-		 (unsigned long long)lba, (unsigned long long)range);
+	struct se_subsystem_dev *su_dev = dev->se_sub_dev;
+	unsigned char *cdb = cmd->t_task_cdb;
 
-	ret = dev->transport->do_discard(dev, lba, range);
-	if (ret < 0) {
-		pr_debug("blkdev_issue_discard() failed for WRITE_SAME\n");
-		return ret;
-	}
+	switch (cdb[0]) {
+	case MODE_SELECT:
+		*size = cdb[4];
+		break;
+	case MODE_SELECT_10:
+		*size = (cdb[7] << 8) + cdb[8];
+		break;
+	case MODE_SENSE:
+		*size = cdb[4];
+		cmd->execute_cmd = spc_emulate_modesense;
+		break;
+	case MODE_SENSE_10:
+		*size = (cdb[7] << 8) + cdb[8];
+		cmd->execute_cmd = spc_emulate_modesense;
+		break;
+	case LOG_SELECT:
+	case LOG_SENSE:
+		*size = (cdb[7] << 8) + cdb[8];
+		break;
+	case PERSISTENT_RESERVE_IN:
+		if (su_dev->t10_pr.res_type == SPC3_PERSISTENT_RESERVATIONS)
+			cmd->execute_cmd = target_scsi3_emulate_pr_in;
+		*size = (cdb[7] << 8) + cdb[8];
+		break;
+	case PERSISTENT_RESERVE_OUT:
+		if (su_dev->t10_pr.res_type == SPC3_PERSISTENT_RESERVATIONS)
+			cmd->execute_cmd = target_scsi3_emulate_pr_out;
+		*size = (cdb[7] << 8) + cdb[8];
+		break;
+	case RELEASE:
+	case RELEASE_10:
+		if (cdb[0] == RELEASE_10)
+			*size = (cdb[7] << 8) | cdb[8];
+		else
+			*size = cmd->data_length;
+
+		if (su_dev->t10_pr.res_type != SPC_PASSTHROUGH)
+			cmd->execute_cmd = target_scsi2_reservation_release;
+		break;
+	case RESERVE:
+	case RESERVE_10:
+		/*
+		 * The SPC-2 RESERVE does not contain a size in the SCSI CDB.
+		 * Assume the passthrough or $FABRIC_MOD will tell us about it.
+		 */
+		if (cdb[0] == RESERVE_10)
+			*size = (cdb[7] << 8) | cdb[8];
+		else
+			*size = cmd->data_length;
 
-	target_complete_cmd(cmd, GOOD);
-	return 0;
-}
+		/*
+		 * Setup the legacy emulated handler for SPC-2 and
+		 * >= SPC-3 compatible reservation handling (CRH=1)
+		 * Otherwise, we assume the underlying SCSI logic is
+		 * is running in SPC_PASSTHROUGH, and wants reservations
+		 * emulation disabled.
+		 */
+		if (su_dev->t10_pr.res_type != SPC_PASSTHROUGH)
+			cmd->execute_cmd = target_scsi2_reservation_reserve;
+		break;
+	case REQUEST_SENSE:
+		*size = cdb[4];
+		cmd->execute_cmd = spc_emulate_request_sense;
+		break;
+	case INQUIRY:
+		*size = (cdb[3] << 8) + cdb[4];
 
-int target_emulate_synchronize_cache(struct se_cmd *cmd)
-{
-	if (!cmd->se_dev->transport->do_sync_cache) {
-		pr_err("SYNCHRONIZE_CACHE emulation not supported"
-			" for: %s\n", cmd->se_dev->transport->name);
+		/*
+		 * Do implict HEAD_OF_QUEUE processing for INQUIRY.
+		 * See spc4r17 section 5.3
+		 */
+		if (cmd->se_dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
+			cmd->sam_task_attr = MSG_HEAD_TAG;
+		cmd->execute_cmd = spc_emulate_inquiry;
+		break;
+	case SECURITY_PROTOCOL_IN:
+	case SECURITY_PROTOCOL_OUT:
+		*size = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
+		break;
+	case EXTENDED_COPY:
+	case READ_ATTRIBUTE:
+	case RECEIVE_COPY_RESULTS:
+	case WRITE_ATTRIBUTE:
+		*size = (cdb[10] << 24) | (cdb[11] << 16) |
+		       (cdb[12] << 8) | cdb[13];
+		break;
+	case RECEIVE_DIAGNOSTIC:
+	case SEND_DIAGNOSTIC:
+		*size = (cdb[3] << 8) | cdb[4];
+		break;
+	case WRITE_BUFFER:
+		*size = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
+		break;
+	case REPORT_LUNS:
+		cmd->execute_cmd = target_report_luns;
+		*size = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
+		/*
+		 * Do implict HEAD_OF_QUEUE processing for REPORT_LUNS
+		 * See spc4r17 section 5.3
+		 */
+		if (cmd->se_dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
+			cmd->sam_task_attr = MSG_HEAD_TAG;
+		break;
+	case TEST_UNIT_READY:
+		cmd->execute_cmd = spc_emulate_testunitready;
+		*size = 0;
+		break;
+	case MAINTENANCE_IN:
+		if (dev->transport->get_device_type(dev) != TYPE_ROM) {
+			/*
+			 * MAINTENANCE_IN from SCC-2
+			 * Check for emulated MI_REPORT_TARGET_PGS
+			 */
+			if ((cdb[1] & 0x1f) == MI_REPORT_TARGET_PGS &&
+			    su_dev->t10_alua.alua_type == SPC3_ALUA_EMULATED) {
+				cmd->execute_cmd =
+					target_emulate_report_target_port_groups;
+			}
+			*size = get_unaligned_be32(&cdb[6]);
+		} else {
+			/*
+			 * GPCMD_SEND_KEY from multi media commands
+			 */
+			*size = get_unaligned_be16(&cdb[8]);
+		}
+		break;
+	case MAINTENANCE_OUT:
+		if (dev->transport->get_device_type(dev) != TYPE_ROM) {
+			/*
+			 * MAINTENANCE_OUT from SCC-2
+			 * Check for emulated MO_SET_TARGET_PGS.
+			 */
+			if (cdb[1] == MO_SET_TARGET_PGS &&
+			    su_dev->t10_alua.alua_type == SPC3_ALUA_EMULATED) {
+				cmd->execute_cmd =
+					target_emulate_set_target_port_groups;
+			}
+			*size = get_unaligned_be32(&cdb[6]);
+		} else {
+			/*
+			 * GPCMD_SEND_KEY from multi media commands
+			 */
+			*size = get_unaligned_be16(&cdb[8]);
+		}
+		break;
+	default:
+		pr_warn("TARGET_CORE[%s]: Unsupported SCSI Opcode"
+			" 0x%02x, sending CHECK_CONDITION.\n",
+			cmd->se_tfo->get_fabric_name(), cdb[0]);
+		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
 		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
-		return -ENOSYS;
+		return -EINVAL;
 	}
 
-	cmd->se_dev->transport->do_sync_cache(cmd);
-	return 0;
-}
-
-int target_emulate_noop(struct se_cmd *cmd)
-{
-	target_complete_cmd(cmd, GOOD);
 	return 0;
 }
+EXPORT_SYMBOL(spc_parse_cdb);
diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c
index 84caf1bed9a3..1c59a3c23b2c 100644
--- a/drivers/target/target_core_tmr.c
+++ b/drivers/target/target_core_tmr.c
@@ -295,9 +295,6 @@ static void core_tmr_drain_state_list(
 
 		list_move_tail(&cmd->state_list, &drain_task_list);
 		cmd->state_active = false;
-
-		if (!list_empty(&cmd->execute_list))
-			__target_remove_from_execute_list(cmd);
 	}
 	spin_unlock_irqrestore(&dev->execute_task_lock, flags);
 
@@ -354,57 +351,6 @@ static void core_tmr_drain_state_list(
 	}
 }
 
-static void core_tmr_drain_cmd_list(
-	struct se_device *dev,
-	struct se_cmd *prout_cmd,
-	struct se_node_acl *tmr_nacl,
-	int tas,
-	struct list_head *preempt_and_abort_list)
-{
-	LIST_HEAD(drain_cmd_list);
-	struct se_queue_obj *qobj = &dev->dev_queue_obj;
-	struct se_cmd *cmd, *tcmd;
-	unsigned long flags;
-
-	/*
-	 * Release all commands remaining in the per-device command queue.
-	 *
-	 * This follows the same logic as above for the state list.
-	 */
-	spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
-	list_for_each_entry_safe(cmd, tcmd, &qobj->qobj_list, se_queue_node) {
-		/*
-		 * For PREEMPT_AND_ABORT usage, only process commands
-		 * with a matching reservation key.
-		 */
-		if (target_check_cdb_and_preempt(preempt_and_abort_list, cmd))
-			continue;
-		/*
-		 * Not aborting PROUT PREEMPT_AND_ABORT CDB..
-		 */
-		if (prout_cmd == cmd)
-			continue;
-
-		cmd->transport_state &= ~CMD_T_QUEUED;
-		atomic_dec(&qobj->queue_cnt);
-		list_move_tail(&cmd->se_queue_node, &drain_cmd_list);
-	}
-	spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
-
-	while (!list_empty(&drain_cmd_list)) {
-		cmd = list_entry(drain_cmd_list.next, struct se_cmd, se_queue_node);
-		list_del_init(&cmd->se_queue_node);
-
-		pr_debug("LUN_RESET: %s from Device Queue: cmd: %p t_state:"
-			" %d t_fe_count: %d\n", (preempt_and_abort_list) ?
-			"Preempt" : "", cmd, cmd->t_state,
-			atomic_read(&cmd->t_fe_count));
-
-		core_tmr_handle_tas_abort(tmr_nacl, cmd, tas,
-				atomic_read(&cmd->t_fe_count));
-	}
-}
-
 int core_tmr_lun_reset(
         struct se_device *dev,
         struct se_tmr_req *tmr,
@@ -447,8 +393,7 @@ int core_tmr_lun_reset(
 	core_tmr_drain_tmr_list(dev, tmr, preempt_and_abort_list);
 	core_tmr_drain_state_list(dev, prout_cmd, tmr_nacl, tas,
 				preempt_and_abort_list);
-	core_tmr_drain_cmd_list(dev, prout_cmd, tmr_nacl, tas,
-				preempt_and_abort_list);
+
 	/*
 	 * Clear any legacy SPC-2 reservation when called during
 	 * LOGICAL UNIT RESET
diff --git a/drivers/target/target_core_tpg.c b/drivers/target/target_core_tpg.c
index 8bd58e284185..b8628a5014b9 100644
--- a/drivers/target/target_core_tpg.c
+++ b/drivers/target/target_core_tpg.c
@@ -77,8 +77,8 @@ static void core_clear_initiator_node_from_tpg(
 
 		lun = deve->se_lun;
 		spin_unlock_irq(&nacl->device_list_lock);
-		core_update_device_list_for_node(lun, NULL, deve->mapped_lun,
-			TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg, 0);
+		core_disable_device_list_for_node(lun, NULL, deve->mapped_lun,
+			TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg);
 
 		spin_lock_irq(&nacl->device_list_lock);
 	}
@@ -172,8 +172,8 @@ void core_tpg_add_node_to_devs(
 			(lun_access == TRANSPORT_LUNFLAGS_READ_WRITE) ?
 			"READ-WRITE" : "READ-ONLY");
 
-		core_update_device_list_for_node(lun, NULL, lun->unpacked_lun,
-				lun_access, acl, tpg, 1);
+		core_enable_device_list_for_node(lun, NULL, lun->unpacked_lun,
+				lun_access, acl, tpg);
 		spin_lock(&tpg->tpg_lun_lock);
 	}
 	spin_unlock(&tpg->tpg_lun_lock);
@@ -306,10 +306,8 @@ struct se_node_acl *core_tpg_check_initiator_node_acl(
 	 * TPG LUNs if the fabric is not explictly asking for
 	 * tpg_check_demo_mode_login_only() == 1.
 	 */
-	if ((tpg->se_tpg_tfo->tpg_check_demo_mode_login_only != NULL) &&
-	    (tpg->se_tpg_tfo->tpg_check_demo_mode_login_only(tpg) == 1))
-		do { ; } while (0);
-	else
+	if ((tpg->se_tpg_tfo->tpg_check_demo_mode_login_only == NULL) ||
+	    (tpg->se_tpg_tfo->tpg_check_demo_mode_login_only(tpg) != 1))
 		core_tpg_add_node_to_devs(acl, tpg);
 
 	spin_lock_irq(&tpg->acl_node_lock);
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 634d0f31a28c..0eaae23d12b5 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -66,15 +66,12 @@ struct kmem_cache *t10_alua_lu_gp_mem_cache;
 struct kmem_cache *t10_alua_tg_pt_gp_cache;
 struct kmem_cache *t10_alua_tg_pt_gp_mem_cache;
 
-static int transport_generic_write_pending(struct se_cmd *);
-static int transport_processing_thread(void *param);
-static int __transport_execute_tasks(struct se_device *dev, struct se_cmd *);
 static void transport_complete_task_attr(struct se_cmd *cmd);
 static void transport_handle_queue_full(struct se_cmd *cmd,
 		struct se_device *dev);
 static int transport_generic_get_mem(struct se_cmd *cmd);
+static int target_get_sess_cmd(struct se_session *, struct se_cmd *, bool);
 static void transport_put_cmd(struct se_cmd *cmd);
-static void transport_remove_cmd_from_queue(struct se_cmd *cmd);
 static int transport_set_sense_codes(struct se_cmd *cmd, u8 asc, u8 ascq);
 static void target_complete_ok_work(struct work_struct *work);
 
@@ -195,14 +192,6 @@ u32 scsi_get_new_index(scsi_index_t type)
 	return new_index;
 }
 
-static void transport_init_queue_obj(struct se_queue_obj *qobj)
-{
-	atomic_set(&qobj->queue_cnt, 0);
-	INIT_LIST_HEAD(&qobj->qobj_list);
-	init_waitqueue_head(&qobj->thread_wq);
-	spin_lock_init(&qobj->cmd_queue_lock);
-}
-
 void transport_subsystem_check_init(void)
 {
 	int ret;
@@ -243,7 +232,6 @@ struct se_session *transport_init_session(void)
 	INIT_LIST_HEAD(&se_sess->sess_list);
 	INIT_LIST_HEAD(&se_sess->sess_acl_list);
 	INIT_LIST_HEAD(&se_sess->sess_cmd_list);
-	INIT_LIST_HEAD(&se_sess->sess_wait_list);
 	spin_lock_init(&se_sess->sess_cmd_lock);
 	kref_init(&se_sess->sess_kref);
 
@@ -468,18 +456,7 @@ static void target_remove_from_state_list(struct se_cmd *cmd)
 	spin_unlock_irqrestore(&dev->execute_task_lock, flags);
 }
 
-/*	transport_cmd_check_stop():
- *
- *	'transport_off = 1' determines if CMD_T_ACTIVE should be cleared.
- *	'transport_off = 2' determines if task_dev_state should be removed.
- *
- *	A non-zero u8 t_state sets cmd->t_state.
- *	Returns 1 when command is stopped, else 0.
- */
-static int transport_cmd_check_stop(
-	struct se_cmd *cmd,
-	int transport_off,
-	u8 t_state)
+static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists)
 {
 	unsigned long flags;
 
@@ -493,13 +470,23 @@ static int transport_cmd_check_stop(
 			__func__, __LINE__, cmd->se_tfo->get_task_tag(cmd));
 
 		cmd->transport_state &= ~CMD_T_ACTIVE;
-		if (transport_off == 2)
+		if (remove_from_lists)
 			target_remove_from_state_list(cmd);
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
 		complete(&cmd->transport_lun_stop_comp);
 		return 1;
 	}
+
+	if (remove_from_lists) {
+		target_remove_from_state_list(cmd);
+
+		/*
+		 * Clear struct se_cmd->se_lun before the handoff to FE.
+		 */
+		cmd->se_lun = NULL;
+	}
+
 	/*
 	 * Determine if frontend context caller is requesting the stopping of
 	 * this command for frontend exceptions.
@@ -509,58 +496,36 @@ static int transport_cmd_check_stop(
 			__func__, __LINE__,
 			cmd->se_tfo->get_task_tag(cmd));
 
-		if (transport_off == 2)
-			target_remove_from_state_list(cmd);
-
-		/*
-		 * Clear struct se_cmd->se_lun before the transport_off == 2 handoff
-		 * to FE.
-		 */
-		if (transport_off == 2)
-			cmd->se_lun = NULL;
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
 		complete(&cmd->t_transport_stop_comp);
 		return 1;
 	}
-	if (transport_off) {
-		cmd->transport_state &= ~CMD_T_ACTIVE;
-		if (transport_off == 2) {
-			target_remove_from_state_list(cmd);
-			/*
-			 * Clear struct se_cmd->se_lun before the transport_off == 2
-			 * handoff to fabric module.
-			 */
-			cmd->se_lun = NULL;
-			/*
-			 * Some fabric modules like tcm_loop can release
-			 * their internally allocated I/O reference now and
-			 * struct se_cmd now.
-			 *
-			 * Fabric modules are expected to return '1' here if the
-			 * se_cmd being passed is released at this point,
-			 * or zero if not being released.
-			 */
-			if (cmd->se_tfo->check_stop_free != NULL) {
-				spin_unlock_irqrestore(
-					&cmd->t_state_lock, flags);
-
-				return cmd->se_tfo->check_stop_free(cmd);
-			}
+
+	cmd->transport_state &= ~CMD_T_ACTIVE;
+	if (remove_from_lists) {
+		/*
+		 * Some fabric modules like tcm_loop can release
+		 * their internally allocated I/O reference now and
+		 * struct se_cmd now.
+		 *
+		 * Fabric modules are expected to return '1' here if the
+		 * se_cmd being passed is released at this point,
+		 * or zero if not being released.
+		 */
+		if (cmd->se_tfo->check_stop_free != NULL) {
+			spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+			return cmd->se_tfo->check_stop_free(cmd);
 		}
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	}
 
-		return 0;
-	} else if (t_state)
-		cmd->t_state = t_state;
 	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-
 	return 0;
 }
 
 static int transport_cmd_check_stop_to_fabric(struct se_cmd *cmd)
 {
-	return transport_cmd_check_stop(cmd, 2, 0);
+	return transport_cmd_check_stop(cmd, true);
 }
 
 static void transport_lun_remove_cmd(struct se_cmd *cmd)
@@ -591,79 +556,8 @@ void transport_cmd_finish_abort(struct se_cmd *cmd, int remove)
 
 	if (transport_cmd_check_stop_to_fabric(cmd))
 		return;
-	if (remove) {
-		transport_remove_cmd_from_queue(cmd);
+	if (remove)
 		transport_put_cmd(cmd);
-	}
-}
-
-static void transport_add_cmd_to_queue(struct se_cmd *cmd, int t_state,
-		bool at_head)
-{
-	struct se_device *dev = cmd->se_dev;
-	struct se_queue_obj *qobj = &dev->dev_queue_obj;
-	unsigned long flags;
-
-	if (t_state) {
-		spin_lock_irqsave(&cmd->t_state_lock, flags);
-		cmd->t_state = t_state;
-		cmd->transport_state |= CMD_T_ACTIVE;
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-	}
-
-	spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
-
-	/* If the cmd is already on the list, remove it before we add it */
-	if (!list_empty(&cmd->se_queue_node))
-		list_del(&cmd->se_queue_node);
-	else
-		atomic_inc(&qobj->queue_cnt);
-
-	if (at_head)
-		list_add(&cmd->se_queue_node, &qobj->qobj_list);
-	else
-		list_add_tail(&cmd->se_queue_node, &qobj->qobj_list);
-	cmd->transport_state |= CMD_T_QUEUED;
-	spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
-
-	wake_up_interruptible(&qobj->thread_wq);
-}
-
-static struct se_cmd *
-transport_get_cmd_from_queue(struct se_queue_obj *qobj)
-{
-	struct se_cmd *cmd;
-	unsigned long flags;
-
-	spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
-	if (list_empty(&qobj->qobj_list)) {
-		spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
-		return NULL;
-	}
-	cmd = list_first_entry(&qobj->qobj_list, struct se_cmd, se_queue_node);
-
-	cmd->transport_state &= ~CMD_T_QUEUED;
-	list_del_init(&cmd->se_queue_node);
-	atomic_dec(&qobj->queue_cnt);
-	spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
-
-	return cmd;
-}
-
-static void transport_remove_cmd_from_queue(struct se_cmd *cmd)
-{
-	struct se_queue_obj *qobj = &cmd->se_dev->dev_queue_obj;
-	unsigned long flags;
-
-	spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
-	if (!(cmd->transport_state & CMD_T_QUEUED)) {
-		spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
-		return;
-	}
-	cmd->transport_state &= ~CMD_T_QUEUED;
-	atomic_dec(&qobj->queue_cnt);
-	list_del_init(&cmd->se_queue_node);
-	spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
 }
 
 static void target_complete_failure_work(struct work_struct *work)
@@ -742,68 +636,11 @@ static void target_add_to_state_list(struct se_cmd *cmd)
 	spin_unlock_irqrestore(&dev->execute_task_lock, flags);
 }
 
-static void __target_add_to_execute_list(struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-	bool head_of_queue = false;
-
-	if (!list_empty(&cmd->execute_list))
-		return;
-
-	if (dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED &&
-	    cmd->sam_task_attr == MSG_HEAD_TAG)
-		head_of_queue = true;
-
-	if (head_of_queue)
-		list_add(&cmd->execute_list, &dev->execute_list);
-	else
-		list_add_tail(&cmd->execute_list, &dev->execute_list);
-
-	atomic_inc(&dev->execute_tasks);
-
-	if (cmd->state_active)
-		return;
-
-	if (head_of_queue)
-		list_add(&cmd->state_list, &dev->state_list);
-	else
-		list_add_tail(&cmd->state_list, &dev->state_list);
-
-	cmd->state_active = true;
-}
-
-static void target_add_to_execute_list(struct se_cmd *cmd)
-{
-	unsigned long flags;
-	struct se_device *dev = cmd->se_dev;
-
-	spin_lock_irqsave(&dev->execute_task_lock, flags);
-	__target_add_to_execute_list(cmd);
-	spin_unlock_irqrestore(&dev->execute_task_lock, flags);
-}
-
-void __target_remove_from_execute_list(struct se_cmd *cmd)
-{
-	list_del_init(&cmd->execute_list);
-	atomic_dec(&cmd->se_dev->execute_tasks);
-}
-
-static void target_remove_from_execute_list(struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-	unsigned long flags;
-
-	if (WARN_ON(list_empty(&cmd->execute_list)))
-		return;
-
-	spin_lock_irqsave(&dev->execute_task_lock, flags);
-	__target_remove_from_execute_list(cmd);
-	spin_unlock_irqrestore(&dev->execute_task_lock, flags);
-}
-
 /*
  * Handle QUEUE_FULL / -EAGAIN and -ENOMEM status
  */
+static void transport_write_pending_qf(struct se_cmd *cmd);
+static void transport_complete_qf(struct se_cmd *cmd);
 
 static void target_qf_do_work(struct work_struct *work)
 {
@@ -827,7 +664,10 @@ static void target_qf_do_work(struct work_struct *work)
 			(cmd->t_state == TRANSPORT_COMPLETE_QF_WP) ? "WRITE_PENDING"
 			: "UNKNOWN");
 
-		transport_add_cmd_to_queue(cmd, cmd->t_state, true);
+		if (cmd->t_state == TRANSPORT_COMPLETE_QF_WP)
+			transport_write_pending_qf(cmd);
+		else if (cmd->t_state == TRANSPORT_COMPLETE_QF_OK)
+			transport_complete_qf(cmd);
 	}
 }
 
@@ -874,8 +714,7 @@ void transport_dump_dev_state(
 		break;
 	}
 
-	*bl += sprintf(b + *bl, "  Execute/Max Queue Depth: %d/%d",
-		atomic_read(&dev->execute_tasks), dev->queue_depth);
+	*bl += sprintf(b + *bl, "  Max Queue Depth: %d", dev->queue_depth);
 	*bl += sprintf(b + *bl, "  SectorSize: %u  HwMaxSectors: %u\n",
 		dev->se_sub_dev->se_dev_attrib.block_size,
 		dev->se_sub_dev->se_dev_attrib.hw_max_sectors);
@@ -1212,7 +1051,6 @@ struct se_device *transport_add_device_to_core_hba(
 		return NULL;
 	}
 
-	transport_init_queue_obj(&dev->dev_queue_obj);
 	dev->dev_flags		= device_flags;
 	dev->dev_status		|= TRANSPORT_DEVICE_DEACTIVATED;
 	dev->dev_ptr		= transport_dev;
@@ -1222,7 +1060,6 @@ struct se_device *transport_add_device_to_core_hba(
 	INIT_LIST_HEAD(&dev->dev_list);
 	INIT_LIST_HEAD(&dev->dev_sep_list);
 	INIT_LIST_HEAD(&dev->dev_tmr_list);
-	INIT_LIST_HEAD(&dev->execute_list);
 	INIT_LIST_HEAD(&dev->delayed_cmd_list);
 	INIT_LIST_HEAD(&dev->state_list);
 	INIT_LIST_HEAD(&dev->qf_cmd_list);
@@ -1261,17 +1098,17 @@ struct se_device *transport_add_device_to_core_hba(
 	 * Setup the Asymmetric Logical Unit Assignment for struct se_device
 	 */
 	if (core_setup_alua(dev, force_pt) < 0)
-		goto out;
+		goto err_dev_list;
 
 	/*
 	 * Startup the struct se_device processing thread
 	 */
-	dev->process_thread = kthread_run(transport_processing_thread, dev,
-					  "LIO_%s", dev->transport->name);
-	if (IS_ERR(dev->process_thread)) {
-		pr_err("Unable to create kthread: LIO_%s\n",
+	dev->tmr_wq = alloc_workqueue("tmr-%s", WQ_MEM_RECLAIM | WQ_UNBOUND, 1,
+				      dev->transport->name);
+	if (!dev->tmr_wq) {
+		pr_err("Unable to create tmr workqueue for %s\n",
 			dev->transport->name);
-		goto out;
+		goto err_dev_list;
 	}
 	/*
 	 * Setup work_queue for QUEUE_FULL
@@ -1289,7 +1126,7 @@ struct se_device *transport_add_device_to_core_hba(
 		if (!inquiry_prod || !inquiry_rev) {
 			pr_err("All non TCM/pSCSI plugins require"
 				" INQUIRY consts\n");
-			goto out;
+			goto err_wq;
 		}
 
 		strncpy(&dev->se_sub_dev->t10_wwn.vendor[0], "LIO-ORG", 8);
@@ -1299,9 +1136,10 @@ struct se_device *transport_add_device_to_core_hba(
 	scsi_dump_inquiry(dev);
 
 	return dev;
-out:
-	kthread_stop(dev->process_thread);
 
+err_wq:
+	destroy_workqueue(dev->tmr_wq);
+err_dev_list:
 	spin_lock(&hba->device_lock);
 	list_del(&dev->dev_list);
 	hba->dev_count--;
@@ -1315,35 +1153,54 @@ out:
 }
 EXPORT_SYMBOL(transport_add_device_to_core_hba);
 
-/*	transport_generic_prepare_cdb():
- *
- *	Since the Initiator sees iSCSI devices as LUNs,  the SCSI CDB will
- *	contain the iSCSI LUN in bits 7-5 of byte 1 as per SAM-2.
- *	The point of this is since we are mapping iSCSI LUNs to
- *	SCSI Target IDs having a non-zero LUN in the CDB will throw the
- *	devices and HBAs for a loop.
- */
-static inline void transport_generic_prepare_cdb(
-	unsigned char *cdb)
+int target_cmd_size_check(struct se_cmd *cmd, unsigned int size)
 {
-	switch (cdb[0]) {
-	case READ_10: /* SBC - RDProtect */
-	case READ_12: /* SBC - RDProtect */
-	case READ_16: /* SBC - RDProtect */
-	case SEND_DIAGNOSTIC: /* SPC - SELF-TEST Code */
-	case VERIFY: /* SBC - VRProtect */
-	case VERIFY_16: /* SBC - VRProtect */
-	case WRITE_VERIFY: /* SBC - VRProtect */
-	case WRITE_VERIFY_12: /* SBC - VRProtect */
-	case MAINTENANCE_IN: /* SPC - Parameter Data Format for SA RTPG */
-		break;
-	default:
-		cdb[1] &= 0x1f; /* clear logical unit number */
-		break;
+	struct se_device *dev = cmd->se_dev;
+
+	if (cmd->unknown_data_length) {
+		cmd->data_length = size;
+	} else if (size != cmd->data_length) {
+		pr_warn("TARGET_CORE[%s]: Expected Transfer Length:"
+			" %u does not match SCSI CDB Length: %u for SAM Opcode:"
+			" 0x%02x\n", cmd->se_tfo->get_fabric_name(),
+				cmd->data_length, size, cmd->t_task_cdb[0]);
+
+		cmd->cmd_spdtl = size;
+
+		if (cmd->data_direction == DMA_TO_DEVICE) {
+			pr_err("Rejecting underflow/overflow"
+					" WRITE data\n");
+			goto out_invalid_cdb_field;
+		}
+		/*
+		 * Reject READ_* or WRITE_* with overflow/underflow for
+		 * type SCF_SCSI_DATA_CDB.
+		 */
+		if (dev->se_sub_dev->se_dev_attrib.block_size != 512)  {
+			pr_err("Failing OVERFLOW/UNDERFLOW for LBA op"
+				" CDB on non 512-byte sector setup subsystem"
+				" plugin: %s\n", dev->transport->name);
+			/* Returns CHECK_CONDITION + INVALID_CDB_FIELD */
+			goto out_invalid_cdb_field;
+		}
+
+		if (size > cmd->data_length) {
+			cmd->se_cmd_flags |= SCF_OVERFLOW_BIT;
+			cmd->residual_count = (size - cmd->data_length);
+		} else {
+			cmd->se_cmd_flags |= SCF_UNDERFLOW_BIT;
+			cmd->residual_count = (cmd->data_length - size);
+		}
+		cmd->data_length = size;
 	}
-}
 
-static int transport_generic_cmd_sequencer(struct se_cmd *, unsigned char *);
+	return 0;
+
+out_invalid_cdb_field:
+	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+	cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+	return -EINVAL;
+}
 
 /*
  * Used by fabric modules containing a local struct se_cmd within their
@@ -1361,9 +1218,7 @@ void transport_init_se_cmd(
 	INIT_LIST_HEAD(&cmd->se_lun_node);
 	INIT_LIST_HEAD(&cmd->se_delayed_node);
 	INIT_LIST_HEAD(&cmd->se_qf_node);
-	INIT_LIST_HEAD(&cmd->se_queue_node);
 	INIT_LIST_HEAD(&cmd->se_cmd_list);
-	INIT_LIST_HEAD(&cmd->execute_list);
 	INIT_LIST_HEAD(&cmd->state_list);
 	init_completion(&cmd->transport_lun_fe_stop_comp);
 	init_completion(&cmd->transport_lun_stop_comp);
@@ -1418,9 +1273,12 @@ int target_setup_cmd_from_cdb(
 	struct se_cmd *cmd,
 	unsigned char *cdb)
 {
+	struct se_subsystem_dev *su_dev = cmd->se_dev->se_sub_dev;
+	u32 pr_reg_type = 0;
+	u8 alua_ascq = 0;
+	unsigned long flags;
 	int ret;
 
-	transport_generic_prepare_cdb(cdb);
 	/*
 	 * Ensure that the received CDB is less than the max (252 + 8) bytes
 	 * for VARIABLE_LENGTH_CMD
@@ -1457,15 +1315,66 @@ int target_setup_cmd_from_cdb(
 	 * Copy the original CDB into cmd->
 	 */
 	memcpy(cmd->t_task_cdb, cdb, scsi_command_size(cdb));
+
 	/*
-	 * Setup the received CDB based on SCSI defined opcodes and
-	 * perform unit attention, persistent reservations and ALUA
-	 * checks for virtual device backends.  The cmd->t_task_cdb
-	 * pointer is expected to be setup before we reach this point.
+	 * Check for an existing UNIT ATTENTION condition
 	 */
-	ret = transport_generic_cmd_sequencer(cmd, cdb);
+	if (core_scsi3_ua_check(cmd, cdb) < 0) {
+		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+		cmd->scsi_sense_reason = TCM_CHECK_CONDITION_UNIT_ATTENTION;
+		return -EINVAL;
+	}
+
+	ret = su_dev->t10_alua.alua_state_check(cmd, cdb, &alua_ascq);
+	if (ret != 0) {
+		/*
+		 * Set SCSI additional sense code (ASC) to 'LUN Not Accessible';
+		 * The ALUA additional sense code qualifier (ASCQ) is determined
+		 * by the ALUA primary or secondary access state..
+		 */
+		if (ret > 0) {
+			pr_debug("[%s]: ALUA TG Port not available, "
+				"SenseKey: NOT_READY, ASC/ASCQ: "
+				"0x04/0x%02x\n",
+				cmd->se_tfo->get_fabric_name(), alua_ascq);
+
+			transport_set_sense_codes(cmd, 0x04, alua_ascq);
+			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+			cmd->scsi_sense_reason = TCM_CHECK_CONDITION_NOT_READY;
+			return -EINVAL;
+		}
+		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
+	}
+
+	/*
+	 * Check status for SPC-3 Persistent Reservations
+	 */
+	if (su_dev->t10_pr.pr_ops.t10_reservation_check(cmd, &pr_reg_type)) {
+		if (su_dev->t10_pr.pr_ops.t10_seq_non_holder(
+					cmd, cdb, pr_reg_type) != 0) {
+			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+			cmd->se_cmd_flags |= SCF_SCSI_RESERVATION_CONFLICT;
+			cmd->scsi_status = SAM_STAT_RESERVATION_CONFLICT;
+			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+			return -EBUSY;
+		}
+		/*
+		 * This means the CDB is allowed for the SCSI Initiator port
+		 * when said port is *NOT* holding the legacy SPC-2 or
+		 * SPC-3 Persistent Reservation.
+		 */
+	}
+
+	ret = cmd->se_dev->transport->parse_cdb(cmd);
 	if (ret < 0)
 		return ret;
+
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	cmd->se_cmd_flags |= SCF_SUPPORTED_SAM_OPCODE;
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+
 	/*
 	 * Check for SAM Task Attribute Emulation
 	 */
@@ -1503,10 +1412,9 @@ int transport_handle_cdb_direct(
 		return -EINVAL;
 	}
 	/*
-	 * Set TRANSPORT_NEW_CMD state and CMD_T_ACTIVE following
-	 * transport_generic_handle_cdb*() -> transport_add_cmd_to_queue()
-	 * in existing usage to ensure that outstanding descriptors are handled
-	 * correctly during shutdown via transport_wait_for_tasks()
+	 * Set TRANSPORT_NEW_CMD state and CMD_T_ACTIVE to ensure that
+	 * outstanding descriptors are handled correctly during shutdown via
+	 * transport_wait_for_tasks()
 	 *
 	 * Also, we don't take cmd->t_state_lock here as we only expect
 	 * this to be called for initial descriptor submission.
@@ -1540,10 +1448,14 @@ EXPORT_SYMBOL(transport_handle_cdb_direct);
  * @data_dir: DMA data direction
  * @flags: flags for command submission from target_sc_flags_tables
  *
+ * Returns non zero to signal active I/O shutdown failure.  All other
+ * setup exceptions will be returned as a SCSI CHECK_CONDITION response,
+ * but still return zero here.
+ *
  * This may only be called from process context, and also currently
  * assumes internal allocation of fabric payload buffer by target-core.
  **/
-void target_submit_cmd(struct se_cmd *se_cmd, struct se_session *se_sess,
+int target_submit_cmd(struct se_cmd *se_cmd, struct se_session *se_sess,
 		unsigned char *cdb, unsigned char *sense, u32 unpacked_lun,
 		u32 data_length, int task_attr, int data_dir, int flags)
 {
@@ -1569,7 +1481,9 @@ void target_submit_cmd(struct se_cmd *se_cmd, struct se_session *se_sess,
 	 * for fabrics using TARGET_SCF_ACK_KREF that expect a second
 	 * kref_put() to happen during fabric packet acknowledgement.
 	 */
-	target_get_sess_cmd(se_sess, se_cmd, (flags & TARGET_SCF_ACK_KREF));
+	rc = target_get_sess_cmd(se_sess, se_cmd, (flags & TARGET_SCF_ACK_KREF));
+	if (rc)
+		return rc;
 	/*
 	 * Signal bidirectional data payloads to target-core
 	 */
@@ -1582,16 +1496,13 @@ void target_submit_cmd(struct se_cmd *se_cmd, struct se_session *se_sess,
 		transport_send_check_condition_and_sense(se_cmd,
 				se_cmd->scsi_sense_reason, 0);
 		target_put_sess_cmd(se_sess, se_cmd);
-		return;
+		return 0;
 	}
-	/*
-	 * Sanitize CDBs via transport_generic_cmd_sequencer() and
-	 * allocate the necessary tasks to complete the received CDB+data
-	 */
+
 	rc = target_setup_cmd_from_cdb(se_cmd, cdb);
 	if (rc != 0) {
 		transport_generic_request_failure(se_cmd);
-		return;
+		return 0;
 	}
 
 	/*
@@ -1600,14 +1511,8 @@ void target_submit_cmd(struct se_cmd *se_cmd, struct se_session *se_sess,
 	 */
 	core_alua_check_nonop_delay(se_cmd);
 
-	/*
-	 * Dispatch se_cmd descriptor to se_lun->lun_se_dev backend
-	 * for immediate execution of READs, otherwise wait for
-	 * transport_generic_handle_data() to be called for WRITEs
-	 * when fabric has filled the incoming buffer.
-	 */
 	transport_handle_cdb_direct(se_cmd);
-	return;
+	return 0;
 }
 EXPORT_SYMBOL(target_submit_cmd);
 
@@ -1662,7 +1567,11 @@ int target_submit_tmr(struct se_cmd *se_cmd, struct se_session *se_sess,
 		se_cmd->se_tmr_req->ref_task_tag = tag;
 
 	/* See target_submit_cmd for commentary */
-	target_get_sess_cmd(se_sess, se_cmd, (flags & TARGET_SCF_ACK_KREF));
+	ret = target_get_sess_cmd(se_sess, se_cmd, (flags & TARGET_SCF_ACK_KREF));
+	if (ret) {
+		core_tmr_release_req(se_cmd->se_tmr_req);
+		return ret;
+	}
 
 	ret = transport_lookup_tmr_lun(se_cmd, unpacked_lun);
 	if (ret) {
@@ -1680,67 +1589,6 @@ int target_submit_tmr(struct se_cmd *se_cmd, struct se_session *se_sess,
 EXPORT_SYMBOL(target_submit_tmr);
 
 /*
- * Used by fabric module frontends defining a TFO->new_cmd_map() caller
- * to  queue up a newly setup se_cmd w/ TRANSPORT_NEW_CMD_MAP in order to
- * complete setup in TCM process context w/ TFO->new_cmd_map().
- */
-int transport_generic_handle_cdb_map(
-	struct se_cmd *cmd)
-{
-	if (!cmd->se_lun) {
-		dump_stack();
-		pr_err("cmd->se_lun is NULL\n");
-		return -EINVAL;
-	}
-
-	transport_add_cmd_to_queue(cmd, TRANSPORT_NEW_CMD_MAP, false);
-	return 0;
-}
-EXPORT_SYMBOL(transport_generic_handle_cdb_map);
-
-/*	transport_generic_handle_data():
- *
- *
- */
-int transport_generic_handle_data(
-	struct se_cmd *cmd)
-{
-	/*
-	 * For the software fabric case, then we assume the nexus is being
-	 * failed/shutdown when signals are pending from the kthread context
-	 * caller, so we return a failure.  For the HW target mode case running
-	 * in interrupt code, the signal_pending() check is skipped.
-	 */
-	if (!in_interrupt() && signal_pending(current))
-		return -EPERM;
-	/*
-	 * If the received CDB has aleady been ABORTED by the generic
-	 * target engine, we now call transport_check_aborted_status()
-	 * to queue any delated TASK_ABORTED status for the received CDB to the
-	 * fabric module as we are expecting no further incoming DATA OUT
-	 * sequences at this point.
-	 */
-	if (transport_check_aborted_status(cmd, 1) != 0)
-		return 0;
-
-	transport_add_cmd_to_queue(cmd, TRANSPORT_PROCESS_WRITE, false);
-	return 0;
-}
-EXPORT_SYMBOL(transport_generic_handle_data);
-
-/*	transport_generic_handle_tmr():
- *
- *
- */
-int transport_generic_handle_tmr(
-	struct se_cmd *cmd)
-{
-	transport_add_cmd_to_queue(cmd, TRANSPORT_PROCESS_TMR, false);
-	return 0;
-}
-EXPORT_SYMBOL(transport_generic_handle_tmr);
-
-/*
  * If the cmd is active, request it to be stopped and sleep until it
  * has completed.
  */
@@ -1797,6 +1645,7 @@ void transport_generic_request_failure(struct se_cmd *cmd)
 	case TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE:
 	case TCM_UNKNOWN_MODE_PAGE:
 	case TCM_WRITE_PROTECTED:
+	case TCM_ADDRESS_OUT_OF_RANGE:
 	case TCM_CHECK_CONDITION_ABORT_CMD:
 	case TCM_CHECK_CONDITION_UNIT_ATTENTION:
 	case TCM_CHECK_CONDITION_NOT_READY:
@@ -1832,13 +1681,7 @@ void transport_generic_request_failure(struct se_cmd *cmd)
 		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
 		break;
 	}
-	/*
-	 * If a fabric does not define a cmd->se_tfo->new_cmd_map caller,
-	 * make the call to transport_send_check_condition_and_sense()
-	 * directly.  Otherwise expect the fabric to make the call to
-	 * transport_send_check_condition_and_sense() after handling
-	 * possible unsoliticied write data payloads.
-	 */
+
 	ret = transport_send_check_condition_and_sense(cmd,
 			cmd->scsi_sense_reason, 0);
 	if (ret == -EAGAIN || ret == -ENOMEM)
@@ -1856,406 +1699,123 @@ queue_full:
 }
 EXPORT_SYMBOL(transport_generic_request_failure);
 
-static inline u32 transport_lba_21(unsigned char *cdb)
+static void __target_execute_cmd(struct se_cmd *cmd)
 {
-	return ((cdb[1] & 0x1f) << 16) | (cdb[2] << 8) | cdb[3];
-}
+	int error = 0;
 
-static inline u32 transport_lba_32(unsigned char *cdb)
-{
-	return (cdb[2] << 24) | (cdb[3] << 16) | (cdb[4] << 8) | cdb[5];
-}
-
-static inline unsigned long long transport_lba_64(unsigned char *cdb)
-{
-	unsigned int __v1, __v2;
-
-	__v1 = (cdb[2] << 24) | (cdb[3] << 16) | (cdb[4] << 8) | cdb[5];
-	__v2 = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
-
-	return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
-}
-
-/*
- * For VARIABLE_LENGTH_CDB w/ 32 byte extended CDBs
- */
-static inline unsigned long long transport_lba_64_ext(unsigned char *cdb)
-{
-	unsigned int __v1, __v2;
-
-	__v1 = (cdb[12] << 24) | (cdb[13] << 16) | (cdb[14] << 8) | cdb[15];
-	__v2 = (cdb[16] << 24) | (cdb[17] << 16) | (cdb[18] << 8) | cdb[19];
-
-	return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
-}
-
-static void transport_set_supported_SAM_opcode(struct se_cmd *se_cmd)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&se_cmd->t_state_lock, flags);
-	se_cmd->se_cmd_flags |= SCF_SUPPORTED_SAM_OPCODE;
-	spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
-}
-
-/*
- * Called from Fabric Module context from transport_execute_tasks()
- *
- * The return of this function determins if the tasks from struct se_cmd
- * get added to the execution queue in transport_execute_tasks(),
- * or are added to the delayed or ordered lists here.
- */
-static inline int transport_execute_task_attr(struct se_cmd *cmd)
-{
-	if (cmd->se_dev->dev_task_attr_type != SAM_TASK_ATTR_EMULATED)
-		return 1;
-	/*
-	 * Check for the existence of HEAD_OF_QUEUE, and if true return 1
-	 * to allow the passed struct se_cmd list of tasks to the front of the list.
-	 */
-	 if (cmd->sam_task_attr == MSG_HEAD_TAG) {
-		pr_debug("Added HEAD_OF_QUEUE for CDB:"
-			" 0x%02x, se_ordered_id: %u\n",
-			cmd->t_task_cdb[0],
-			cmd->se_ordered_id);
-		return 1;
-	} else if (cmd->sam_task_attr == MSG_ORDERED_TAG) {
-		atomic_inc(&cmd->se_dev->dev_ordered_sync);
-		smp_mb__after_atomic_inc();
-
-		pr_debug("Added ORDERED for CDB: 0x%02x to ordered"
-				" list, se_ordered_id: %u\n",
-				cmd->t_task_cdb[0],
-				cmd->se_ordered_id);
-		/*
-		 * Add ORDERED command to tail of execution queue if
-		 * no other older commands exist that need to be
-		 * completed first.
-		 */
-		if (!atomic_read(&cmd->se_dev->simple_cmds))
-			return 1;
-	} else {
-		/*
-		 * For SIMPLE and UNTAGGED Task Attribute commands
-		 */
-		atomic_inc(&cmd->se_dev->simple_cmds);
-		smp_mb__after_atomic_inc();
-	}
-	/*
-	 * Otherwise if one or more outstanding ORDERED task attribute exist,
-	 * add the dormant task(s) built for the passed struct se_cmd to the
-	 * execution queue and become in Active state for this struct se_device.
-	 */
-	if (atomic_read(&cmd->se_dev->dev_ordered_sync) != 0) {
-		/*
-		 * Otherwise, add cmd w/ tasks to delayed cmd queue that
-		 * will be drained upon completion of HEAD_OF_QUEUE task.
-		 */
-		spin_lock(&cmd->se_dev->delayed_cmd_lock);
-		cmd->se_cmd_flags |= SCF_DELAYED_CMD_FROM_SAM_ATTR;
-		list_add_tail(&cmd->se_delayed_node,
-				&cmd->se_dev->delayed_cmd_list);
-		spin_unlock(&cmd->se_dev->delayed_cmd_lock);
-
-		pr_debug("Added CDB: 0x%02x Task Attr: 0x%02x to"
-			" delayed CMD list, se_ordered_id: %u\n",
-			cmd->t_task_cdb[0], cmd->sam_task_attr,
-			cmd->se_ordered_id);
-		/*
-		 * Return zero to let transport_execute_tasks() know
-		 * not to add the delayed tasks to the execution list.
-		 */
-		return 0;
-	}
-	/*
-	 * Otherwise, no ORDERED task attributes exist..
-	 */
-	return 1;
-}
-
-/*
- * Called from fabric module context in transport_generic_new_cmd() and
- * transport_generic_process_write()
- */
-static void transport_execute_tasks(struct se_cmd *cmd)
-{
-	int add_tasks;
-	struct se_device *se_dev = cmd->se_dev;
-	/*
-	 * Call transport_cmd_check_stop() to see if a fabric exception
-	 * has occurred that prevents execution.
-	 */
-	if (!transport_cmd_check_stop(cmd, 0, TRANSPORT_PROCESSING)) {
-		/*
-		 * Check for SAM Task Attribute emulation and HEAD_OF_QUEUE
-		 * attribute for the tasks of the received struct se_cmd CDB
-		 */
-		add_tasks = transport_execute_task_attr(cmd);
-		if (add_tasks) {
-			__transport_execute_tasks(se_dev, cmd);
-			return;
-		}
-	}
-	__transport_execute_tasks(se_dev, NULL);
-}
-
-static int __transport_execute_tasks(struct se_device *dev, struct se_cmd *new_cmd)
-{
-	int error;
-	struct se_cmd *cmd = NULL;
-	unsigned long flags;
-
-check_depth:
-	spin_lock_irq(&dev->execute_task_lock);
-	if (new_cmd != NULL)
-		__target_add_to_execute_list(new_cmd);
-
-	if (list_empty(&dev->execute_list)) {
-		spin_unlock_irq(&dev->execute_task_lock);
-		return 0;
-	}
-	cmd = list_first_entry(&dev->execute_list, struct se_cmd, execute_list);
-	__target_remove_from_execute_list(cmd);
-	spin_unlock_irq(&dev->execute_task_lock);
-
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
-	cmd->transport_state |= CMD_T_BUSY;
-	cmd->transport_state |= CMD_T_SENT;
-
-	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	spin_lock_irq(&cmd->t_state_lock);
+	cmd->transport_state |= (CMD_T_BUSY|CMD_T_SENT);
+	spin_unlock_irq(&cmd->t_state_lock);
 
 	if (cmd->execute_cmd)
 		error = cmd->execute_cmd(cmd);
-	else {
-		error = dev->transport->execute_cmd(cmd, cmd->t_data_sg,
-				cmd->t_data_nents, cmd->data_direction);
-	}
 
-	if (error != 0) {
-		spin_lock_irqsave(&cmd->t_state_lock, flags);
-		cmd->transport_state &= ~CMD_T_BUSY;
-		cmd->transport_state &= ~CMD_T_SENT;
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	if (error) {
+		spin_lock_irq(&cmd->t_state_lock);
+		cmd->transport_state &= ~(CMD_T_BUSY|CMD_T_SENT);
+		spin_unlock_irq(&cmd->t_state_lock);
 
 		transport_generic_request_failure(cmd);
 	}
-
-	new_cmd = NULL;
-	goto check_depth;
-
-	return 0;
 }
 
-static inline u32 transport_get_sectors_6(
-	unsigned char *cdb,
-	struct se_cmd *cmd,
-	int *ret)
+void target_execute_cmd(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
 
 	/*
-	 * Assume TYPE_DISK for non struct se_device objects.
-	 * Use 8-bit sector value.
-	 */
-	if (!dev)
-		goto type_disk;
-
-	/*
-	 * Use 24-bit allocation length for TYPE_TAPE.
+	 * If the received CDB has aleady been aborted stop processing it here.
 	 */
-	if (dev->transport->get_device_type(dev) == TYPE_TAPE)
-		return (u32)(cdb[2] << 16) + (cdb[3] << 8) + cdb[4];
-
-	/*
-	 * Everything else assume TYPE_DISK Sector CDB location.
-	 * Use 8-bit sector value.  SBC-3 says:
-	 *
-	 *   A TRANSFER LENGTH field set to zero specifies that 256
-	 *   logical blocks shall be written.  Any other value
-	 *   specifies the number of logical blocks that shall be
-	 *   written.
-	 */
-type_disk:
-	return cdb[4] ? : 256;
-}
-
-static inline u32 transport_get_sectors_10(
-	unsigned char *cdb,
-	struct se_cmd *cmd,
-	int *ret)
-{
-	struct se_device *dev = cmd->se_dev;
+	if (transport_check_aborted_status(cmd, 1))
+		return;
 
 	/*
-	 * Assume TYPE_DISK for non struct se_device objects.
-	 * Use 16-bit sector value.
+	 * Determine if IOCTL context caller in requesting the stopping of this
+	 * command for LUN shutdown purposes.
 	 */
-	if (!dev)
-		goto type_disk;
+	spin_lock_irq(&cmd->t_state_lock);
+	if (cmd->transport_state & CMD_T_LUN_STOP) {
+		pr_debug("%s:%d CMD_T_LUN_STOP for ITT: 0x%08x\n",
+			__func__, __LINE__, cmd->se_tfo->get_task_tag(cmd));
 
-	/*
-	 * XXX_10 is not defined in SSC, throw an exception
-	 */
-	if (dev->transport->get_device_type(dev) == TYPE_TAPE) {
-		*ret = -EINVAL;
-		return 0;
+		cmd->transport_state &= ~CMD_T_ACTIVE;
+		spin_unlock_irq(&cmd->t_state_lock);
+		complete(&cmd->transport_lun_stop_comp);
+		return;
 	}
-
 	/*
-	 * Everything else assume TYPE_DISK Sector CDB location.
-	 * Use 16-bit sector value.
-	 */
-type_disk:
-	return (u32)(cdb[7] << 8) + cdb[8];
-}
-
-static inline u32 transport_get_sectors_12(
-	unsigned char *cdb,
-	struct se_cmd *cmd,
-	int *ret)
-{
-	struct se_device *dev = cmd->se_dev;
-
-	/*
-	 * Assume TYPE_DISK for non struct se_device objects.
-	 * Use 32-bit sector value.
+	 * Determine if frontend context caller is requesting the stopping of
+	 * this command for frontend exceptions.
 	 */
-	if (!dev)
-		goto type_disk;
+	if (cmd->transport_state & CMD_T_STOP) {
+		pr_debug("%s:%d CMD_T_STOP for ITT: 0x%08x\n",
+			__func__, __LINE__,
+			cmd->se_tfo->get_task_tag(cmd));
 
-	/*
-	 * XXX_12 is not defined in SSC, throw an exception
-	 */
-	if (dev->transport->get_device_type(dev) == TYPE_TAPE) {
-		*ret = -EINVAL;
-		return 0;
+		spin_unlock_irq(&cmd->t_state_lock);
+		complete(&cmd->t_transport_stop_comp);
+		return;
 	}
 
-	/*
-	 * Everything else assume TYPE_DISK Sector CDB location.
-	 * Use 32-bit sector value.
-	 */
-type_disk:
-	return (u32)(cdb[6] << 24) + (cdb[7] << 16) + (cdb[8] << 8) + cdb[9];
-}
-
-static inline u32 transport_get_sectors_16(
-	unsigned char *cdb,
-	struct se_cmd *cmd,
-	int *ret)
-{
-	struct se_device *dev = cmd->se_dev;
-
-	/*
-	 * Assume TYPE_DISK for non struct se_device objects.
-	 * Use 32-bit sector value.
-	 */
-	if (!dev)
-		goto type_disk;
+	cmd->t_state = TRANSPORT_PROCESSING;
+	spin_unlock_irq(&cmd->t_state_lock);
 
-	/*
-	 * Use 24-bit allocation length for TYPE_TAPE.
-	 */
-	if (dev->transport->get_device_type(dev) == TYPE_TAPE)
-		return (u32)(cdb[12] << 16) + (cdb[13] << 8) + cdb[14];
-
-type_disk:
-	return (u32)(cdb[10] << 24) + (cdb[11] << 16) +
-		    (cdb[12] << 8) + cdb[13];
-}
+	if (dev->dev_task_attr_type != SAM_TASK_ATTR_EMULATED)
+		goto execute;
 
-/*
- * Used for VARIABLE_LENGTH_CDB WRITE_32 and READ_32 variants
- */
-static inline u32 transport_get_sectors_32(
-	unsigned char *cdb,
-	struct se_cmd *cmd,
-	int *ret)
-{
 	/*
-	 * Assume TYPE_DISK for non struct se_device objects.
-	 * Use 32-bit sector value.
+	 * Check for the existence of HEAD_OF_QUEUE, and if true return 1
+	 * to allow the passed struct se_cmd list of tasks to the front of the list.
 	 */
-	return (u32)(cdb[28] << 24) + (cdb[29] << 16) +
-		    (cdb[30] << 8) + cdb[31];
+	switch (cmd->sam_task_attr) {
+	case MSG_HEAD_TAG:
+		pr_debug("Added HEAD_OF_QUEUE for CDB: 0x%02x, "
+			 "se_ordered_id: %u\n",
+			 cmd->t_task_cdb[0], cmd->se_ordered_id);
+		goto execute;
+	case MSG_ORDERED_TAG:
+		atomic_inc(&dev->dev_ordered_sync);
+		smp_mb__after_atomic_inc();
 
-}
+		pr_debug("Added ORDERED for CDB: 0x%02x to ordered list, "
+			 " se_ordered_id: %u\n",
+			 cmd->t_task_cdb[0], cmd->se_ordered_id);
 
-static inline u32 transport_get_size(
-	u32 sectors,
-	unsigned char *cdb,
-	struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-
-	if (dev->transport->get_device_type(dev) == TYPE_TAPE) {
-		if (cdb[1] & 1) { /* sectors */
-			return dev->se_sub_dev->se_dev_attrib.block_size * sectors;
-		} else /* bytes */
-			return sectors;
+		/*
+		 * Execute an ORDERED command if no other older commands
+		 * exist that need to be completed first.
+		 */
+		if (!atomic_read(&dev->simple_cmds))
+			goto execute;
+		break;
+	default:
+		/*
+		 * For SIMPLE and UNTAGGED Task Attribute commands
+		 */
+		atomic_inc(&dev->simple_cmds);
+		smp_mb__after_atomic_inc();
+		break;
 	}
 
-	pr_debug("Returning block_size: %u, sectors: %u == %u for"
-		" %s object\n", dev->se_sub_dev->se_dev_attrib.block_size,
-		sectors, dev->se_sub_dev->se_dev_attrib.block_size * sectors,
-		dev->transport->name);
-
-	return dev->se_sub_dev->se_dev_attrib.block_size * sectors;
-}
+	if (atomic_read(&dev->dev_ordered_sync) != 0) {
+		spin_lock(&dev->delayed_cmd_lock);
+		list_add_tail(&cmd->se_delayed_node, &dev->delayed_cmd_list);
+		spin_unlock(&dev->delayed_cmd_lock);
 
-static void transport_xor_callback(struct se_cmd *cmd)
-{
-	unsigned char *buf, *addr;
-	struct scatterlist *sg;
-	unsigned int offset;
-	int i;
-	int count;
-	/*
-	 * From sbc3r22.pdf section 5.48 XDWRITEREAD (10) command
-	 *
-	 * 1) read the specified logical block(s);
-	 * 2) transfer logical blocks from the data-out buffer;
-	 * 3) XOR the logical blocks transferred from the data-out buffer with
-	 *    the logical blocks read, storing the resulting XOR data in a buffer;
-	 * 4) if the DISABLE WRITE bit is set to zero, then write the logical
-	 *    blocks transferred from the data-out buffer; and
-	 * 5) transfer the resulting XOR data to the data-in buffer.
-	 */
-	buf = kmalloc(cmd->data_length, GFP_KERNEL);
-	if (!buf) {
-		pr_err("Unable to allocate xor_callback buf\n");
+		pr_debug("Added CDB: 0x%02x Task Attr: 0x%02x to"
+			" delayed CMD list, se_ordered_id: %u\n",
+			cmd->t_task_cdb[0], cmd->sam_task_attr,
+			cmd->se_ordered_id);
 		return;
 	}
-	/*
-	 * Copy the scatterlist WRITE buffer located at cmd->t_data_sg
-	 * into the locally allocated *buf
-	 */
-	sg_copy_to_buffer(cmd->t_data_sg,
-			  cmd->t_data_nents,
-			  buf,
-			  cmd->data_length);
 
+execute:
 	/*
-	 * Now perform the XOR against the BIDI read memory located at
-	 * cmd->t_mem_bidi_list
+	 * Otherwise, no ORDERED task attributes exist..
 	 */
-
-	offset = 0;
-	for_each_sg(cmd->t_bidi_data_sg, sg, cmd->t_bidi_data_nents, count) {
-		addr = kmap_atomic(sg_page(sg));
-		if (!addr)
-			goto out;
-
-		for (i = 0; i < sg->length; i++)
-			*(addr + sg->offset + i) ^= *(buf + offset + i);
-
-		offset += sg->length;
-		kunmap_atomic(addr);
-	}
-
-out:
-	kfree(buf);
+	__target_execute_cmd(cmd);
 }
+EXPORT_SYMBOL(target_execute_cmd);
 
 /*
  * Used to obtain Sense Data from underlying Linux/SCSI struct scsi_cmnd
@@ -2312,737 +1872,31 @@ out:
 	return -1;
 }
 
-static inline long long transport_dev_end_lba(struct se_device *dev)
-{
-	return dev->transport->get_blocks(dev) + 1;
-}
-
-static int transport_cmd_get_valid_sectors(struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-	u32 sectors;
-
-	if (dev->transport->get_device_type(dev) != TYPE_DISK)
-		return 0;
-
-	sectors = (cmd->data_length / dev->se_sub_dev->se_dev_attrib.block_size);
-
-	if ((cmd->t_task_lba + sectors) > transport_dev_end_lba(dev)) {
-		pr_err("LBA: %llu Sectors: %u exceeds"
-			" transport_dev_end_lba(): %llu\n",
-			cmd->t_task_lba, sectors,
-			transport_dev_end_lba(dev));
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int target_check_write_same_discard(unsigned char *flags, struct se_device *dev)
-{
-	/*
-	 * Determine if the received WRITE_SAME is used to for direct
-	 * passthrough into Linux/SCSI with struct request via TCM/pSCSI
-	 * or we are signaling the use of internal WRITE_SAME + UNMAP=1
-	 * emulation for -> Linux/BLOCK disbard with TCM/IBLOCK code.
-	 */
-	int passthrough = (dev->transport->transport_type ==
-				TRANSPORT_PLUGIN_PHBA_PDEV);
-
-	if (!passthrough) {
-		if ((flags[0] & 0x04) || (flags[0] & 0x02)) {
-			pr_err("WRITE_SAME PBDATA and LBDATA"
-				" bits not supported for Block Discard"
-				" Emulation\n");
-			return -ENOSYS;
-		}
-		/*
-		 * Currently for the emulated case we only accept
-		 * tpws with the UNMAP=1 bit set.
-		 */
-		if (!(flags[0] & 0x08)) {
-			pr_err("WRITE_SAME w/o UNMAP bit not"
-				" supported for Block Discard Emulation\n");
-			return -ENOSYS;
-		}
-	}
-
-	return 0;
-}
-
-/*	transport_generic_cmd_sequencer():
- *
- *	Generic Command Sequencer that should work for most DAS transport
- *	drivers.
- *
- *	Called from target_setup_cmd_from_cdb() in the $FABRIC_MOD
- *	RX Thread.
- *
- *	FIXME: Need to support other SCSI OPCODES where as well.
+/*
+ * Process all commands up to the last received ORDERED task attribute which
+ * requires another blocking boundary
  */
-static int transport_generic_cmd_sequencer(
-	struct se_cmd *cmd,
-	unsigned char *cdb)
+static void target_restart_delayed_cmds(struct se_device *dev)
 {
-	struct se_device *dev = cmd->se_dev;
-	struct se_subsystem_dev *su_dev = dev->se_sub_dev;
-	int ret = 0, sector_ret = 0, passthrough;
-	u32 sectors = 0, size = 0, pr_reg_type = 0;
-	u16 service_action;
-	u8 alua_ascq = 0;
-	/*
-	 * Check for an existing UNIT ATTENTION condition
-	 */
-	if (core_scsi3_ua_check(cmd, cdb) < 0) {
-		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-		cmd->scsi_sense_reason = TCM_CHECK_CONDITION_UNIT_ATTENTION;
-		return -EINVAL;
-	}
-	/*
-	 * Check status of Asymmetric Logical Unit Assignment port
-	 */
-	ret = su_dev->t10_alua.alua_state_check(cmd, cdb, &alua_ascq);
-	if (ret != 0) {
-		/*
-		 * Set SCSI additional sense code (ASC) to 'LUN Not Accessible';
-		 * The ALUA additional sense code qualifier (ASCQ) is determined
-		 * by the ALUA primary or secondary access state..
-		 */
-		if (ret > 0) {
-			pr_debug("[%s]: ALUA TG Port not available,"
-				" SenseKey: NOT_READY, ASC/ASCQ: 0x04/0x%02x\n",
-				cmd->se_tfo->get_fabric_name(), alua_ascq);
+	for (;;) {
+		struct se_cmd *cmd;
 
-			transport_set_sense_codes(cmd, 0x04, alua_ascq);
-			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-			cmd->scsi_sense_reason = TCM_CHECK_CONDITION_NOT_READY;
-			return -EINVAL;
-		}
-		goto out_invalid_cdb_field;
-	}
-	/*
-	 * Check status for SPC-3 Persistent Reservations
-	 */
-	if (su_dev->t10_pr.pr_ops.t10_reservation_check(cmd, &pr_reg_type) != 0) {
-		if (su_dev->t10_pr.pr_ops.t10_seq_non_holder(
-					cmd, cdb, pr_reg_type) != 0) {
-			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-			cmd->se_cmd_flags |= SCF_SCSI_RESERVATION_CONFLICT;
-			cmd->scsi_status = SAM_STAT_RESERVATION_CONFLICT;
-			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
-			return -EBUSY;
-		}
-		/*
-		 * This means the CDB is allowed for the SCSI Initiator port
-		 * when said port is *NOT* holding the legacy SPC-2 or
-		 * SPC-3 Persistent Reservation.
-		 */
-	}
-
-	/*
-	 * If we operate in passthrough mode we skip most CDB emulation and
-	 * instead hand the commands down to the physical SCSI device.
-	 */
-	passthrough =
-		(dev->transport->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV);
-
-	switch (cdb[0]) {
-	case READ_6:
-		sectors = transport_get_sectors_6(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_21(cdb);
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case READ_10:
-		sectors = transport_get_sectors_10(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_32(cdb);
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case READ_12:
-		sectors = transport_get_sectors_12(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_32(cdb);
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case READ_16:
-		sectors = transport_get_sectors_16(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_64(cdb);
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case WRITE_6:
-		sectors = transport_get_sectors_6(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_21(cdb);
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case WRITE_10:
-	case WRITE_VERIFY:
-		sectors = transport_get_sectors_10(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_32(cdb);
-		if (cdb[1] & 0x8)
-			cmd->se_cmd_flags |= SCF_FUA;
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case WRITE_12:
-		sectors = transport_get_sectors_12(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_32(cdb);
-		if (cdb[1] & 0x8)
-			cmd->se_cmd_flags |= SCF_FUA;
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case WRITE_16:
-		sectors = transport_get_sectors_16(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_64(cdb);
-		if (cdb[1] & 0x8)
-			cmd->se_cmd_flags |= SCF_FUA;
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-		break;
-	case XDWRITEREAD_10:
-		if ((cmd->data_direction != DMA_TO_DEVICE) ||
-		    !(cmd->se_cmd_flags & SCF_BIDI))
-			goto out_invalid_cdb_field;
-		sectors = transport_get_sectors_10(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->t_task_lba = transport_lba_32(cdb);
-		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-
-		/*
-		 * Do now allow BIDI commands for passthrough mode.
-		 */
-		if (passthrough)
-			goto out_unsupported_cdb;
-
-		/*
-		 * Setup BIDI XOR callback to be run after I/O completion.
-		 */
-		cmd->transport_complete_callback = &transport_xor_callback;
-		if (cdb[1] & 0x8)
-			cmd->se_cmd_flags |= SCF_FUA;
-		break;
-	case VARIABLE_LENGTH_CMD:
-		service_action = get_unaligned_be16(&cdb[8]);
-		switch (service_action) {
-		case XDWRITEREAD_32:
-			sectors = transport_get_sectors_32(cdb, cmd, &sector_ret);
-			if (sector_ret)
-				goto out_unsupported_cdb;
-			size = transport_get_size(sectors, cdb, cmd);
-			/*
-			 * Use WRITE_32 and READ_32 opcodes for the emulated
-			 * XDWRITE_READ_32 logic.
-			 */
-			cmd->t_task_lba = transport_lba_64_ext(cdb);
-			cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
-
-			/*
-			 * Do now allow BIDI commands for passthrough mode.
-			 */
-			if (passthrough)
-				goto out_unsupported_cdb;
-
-			/*
-			 * Setup BIDI XOR callback to be run during after I/O
-			 * completion.
-			 */
-			cmd->transport_complete_callback = &transport_xor_callback;
-			if (cdb[1] & 0x8)
-				cmd->se_cmd_flags |= SCF_FUA;
-			break;
-		case WRITE_SAME_32:
-			sectors = transport_get_sectors_32(cdb, cmd, &sector_ret);
-			if (sector_ret)
-				goto out_unsupported_cdb;
-
-			if (sectors)
-				size = transport_get_size(1, cdb, cmd);
-			else {
-				pr_err("WSNZ=1, WRITE_SAME w/sectors=0 not"
-				       " supported\n");
-				goto out_invalid_cdb_field;
-			}
-
-			cmd->t_task_lba = get_unaligned_be64(&cdb[12]);
-			cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-
-			if (target_check_write_same_discard(&cdb[10], dev) < 0)
-				goto out_unsupported_cdb;
-			if (!passthrough)
-				cmd->execute_cmd = target_emulate_write_same;
-			break;
-		default:
-			pr_err("VARIABLE_LENGTH_CMD service action"
-				" 0x%04x not supported\n", service_action);
-			goto out_unsupported_cdb;
-		}
-		break;
-	case MAINTENANCE_IN:
-		if (dev->transport->get_device_type(dev) != TYPE_ROM) {
-			/* MAINTENANCE_IN from SCC-2 */
-			/*
-			 * Check for emulated MI_REPORT_TARGET_PGS.
-			 */
-			if ((cdb[1] & 0x1f) == MI_REPORT_TARGET_PGS &&
-			    su_dev->t10_alua.alua_type == SPC3_ALUA_EMULATED) {
-				cmd->execute_cmd =
-					target_emulate_report_target_port_groups;
-			}
-			size = (cdb[6] << 24) | (cdb[7] << 16) |
-			       (cdb[8] << 8) | cdb[9];
-		} else {
-			/* GPCMD_SEND_KEY from multi media commands */
-			size = (cdb[8] << 8) + cdb[9];
-		}
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case MODE_SELECT:
-		size = cdb[4];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case MODE_SELECT_10:
-		size = (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case MODE_SENSE:
-		size = cdb[4];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_modesense;
-		break;
-	case MODE_SENSE_10:
-		size = (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_modesense;
-		break;
-	case GPCMD_READ_BUFFER_CAPACITY:
-	case GPCMD_SEND_OPC:
-	case LOG_SELECT:
-	case LOG_SENSE:
-		size = (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case READ_BLOCK_LIMITS:
-		size = READ_BLOCK_LEN;
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case GPCMD_GET_CONFIGURATION:
-	case GPCMD_READ_FORMAT_CAPACITIES:
-	case GPCMD_READ_DISC_INFO:
-	case GPCMD_READ_TRACK_RZONE_INFO:
-		size = (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case PERSISTENT_RESERVE_IN:
-		if (su_dev->t10_pr.res_type == SPC3_PERSISTENT_RESERVATIONS)
-			cmd->execute_cmd = target_scsi3_emulate_pr_in;
-		size = (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case PERSISTENT_RESERVE_OUT:
-		if (su_dev->t10_pr.res_type == SPC3_PERSISTENT_RESERVATIONS)
-			cmd->execute_cmd = target_scsi3_emulate_pr_out;
-		size = (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case GPCMD_MECHANISM_STATUS:
-	case GPCMD_READ_DVD_STRUCTURE:
-		size = (cdb[8] << 8) + cdb[9];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case READ_POSITION:
-		size = READ_POSITION_LEN;
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case MAINTENANCE_OUT:
-		if (dev->transport->get_device_type(dev) != TYPE_ROM) {
-			/* MAINTENANCE_OUT from SCC-2
-			 *
-			 * Check for emulated MO_SET_TARGET_PGS.
-			 */
-			if (cdb[1] == MO_SET_TARGET_PGS &&
-			    su_dev->t10_alua.alua_type == SPC3_ALUA_EMULATED) {
-				cmd->execute_cmd =
-					target_emulate_set_target_port_groups;
-			}
-
-			size = (cdb[6] << 24) | (cdb[7] << 16) |
-			       (cdb[8] << 8) | cdb[9];
-		} else  {
-			/* GPCMD_REPORT_KEY from multi media commands */
-			size = (cdb[8] << 8) + cdb[9];
-		}
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case INQUIRY:
-		size = (cdb[3] << 8) + cdb[4];
-		/*
-		 * Do implict HEAD_OF_QUEUE processing for INQUIRY.
-		 * See spc4r17 section 5.3
-		 */
-		if (cmd->se_dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
-			cmd->sam_task_attr = MSG_HEAD_TAG;
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_inquiry;
-		break;
-	case READ_BUFFER:
-		size = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case READ_CAPACITY:
-		size = READ_CAP_LEN;
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_readcapacity;
-		break;
-	case READ_MEDIA_SERIAL_NUMBER:
-	case SECURITY_PROTOCOL_IN:
-	case SECURITY_PROTOCOL_OUT:
-		size = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case SERVICE_ACTION_IN:
-		switch (cmd->t_task_cdb[1] & 0x1f) {
-		case SAI_READ_CAPACITY_16:
-			if (!passthrough)
-				cmd->execute_cmd =
-					target_emulate_readcapacity_16;
-			break;
-		default:
-			if (passthrough)
-				break;
-
-			pr_err("Unsupported SA: 0x%02x\n",
-				cmd->t_task_cdb[1] & 0x1f);
-			goto out_invalid_cdb_field;
-		}
-		/*FALLTHROUGH*/
-	case ACCESS_CONTROL_IN:
-	case ACCESS_CONTROL_OUT:
-	case EXTENDED_COPY:
-	case READ_ATTRIBUTE:
-	case RECEIVE_COPY_RESULTS:
-	case WRITE_ATTRIBUTE:
-		size = (cdb[10] << 24) | (cdb[11] << 16) |
-		       (cdb[12] << 8) | cdb[13];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case RECEIVE_DIAGNOSTIC:
-	case SEND_DIAGNOSTIC:
-		size = (cdb[3] << 8) | cdb[4];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-/* #warning FIXME: Figure out correct GPCMD_READ_CD blocksize. */
-#if 0
-	case GPCMD_READ_CD:
-		sectors = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
-		size = (2336 * sectors);
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-#endif
-	case READ_TOC:
-		size = cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case REQUEST_SENSE:
-		size = cdb[4];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_request_sense;
-		break;
-	case READ_ELEMENT_STATUS:
-		size = 65536 * cdb[7] + 256 * cdb[8] + cdb[9];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case WRITE_BUFFER:
-		size = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case RESERVE:
-	case RESERVE_10:
-		/*
-		 * The SPC-2 RESERVE does not contain a size in the SCSI CDB.
-		 * Assume the passthrough or $FABRIC_MOD will tell us about it.
-		 */
-		if (cdb[0] == RESERVE_10)
-			size = (cdb[7] << 8) | cdb[8];
-		else
-			size = cmd->data_length;
-
-		/*
-		 * Setup the legacy emulated handler for SPC-2 and
-		 * >= SPC-3 compatible reservation handling (CRH=1)
-		 * Otherwise, we assume the underlying SCSI logic is
-		 * is running in SPC_PASSTHROUGH, and wants reservations
-		 * emulation disabled.
-		 */
-		if (su_dev->t10_pr.res_type != SPC_PASSTHROUGH)
-			cmd->execute_cmd = target_scsi2_reservation_reserve;
-		cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
-		break;
-	case RELEASE:
-	case RELEASE_10:
-		/*
-		 * The SPC-2 RELEASE does not contain a size in the SCSI CDB.
-		 * Assume the passthrough or $FABRIC_MOD will tell us about it.
-		*/
-		if (cdb[0] == RELEASE_10)
-			size = (cdb[7] << 8) | cdb[8];
-		else
-			size = cmd->data_length;
-
-		if (su_dev->t10_pr.res_type != SPC_PASSTHROUGH)
-			cmd->execute_cmd = target_scsi2_reservation_release;
-		cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
-		break;
-	case SYNCHRONIZE_CACHE:
-	case SYNCHRONIZE_CACHE_16:
-		/*
-		 * Extract LBA and range to be flushed for emulated SYNCHRONIZE_CACHE
-		 */
-		if (cdb[0] == SYNCHRONIZE_CACHE) {
-			sectors = transport_get_sectors_10(cdb, cmd, &sector_ret);
-			cmd->t_task_lba = transport_lba_32(cdb);
-		} else {
-			sectors = transport_get_sectors_16(cdb, cmd, &sector_ret);
-			cmd->t_task_lba = transport_lba_64(cdb);
-		}
-		if (sector_ret)
-			goto out_unsupported_cdb;
-
-		size = transport_get_size(sectors, cdb, cmd);
-		cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
-
-		if (passthrough)
+		spin_lock(&dev->delayed_cmd_lock);
+		if (list_empty(&dev->delayed_cmd_list)) {
+			spin_unlock(&dev->delayed_cmd_lock);
 			break;
-
-		/*
-		 * Check to ensure that LBA + Range does not exceed past end of
-		 * device for IBLOCK and FILEIO ->do_sync_cache() backend calls
-		 */
-		if ((cmd->t_task_lba != 0) || (sectors != 0)) {
-			if (transport_cmd_get_valid_sectors(cmd) < 0)
-				goto out_invalid_cdb_field;
-		}
-		cmd->execute_cmd = target_emulate_synchronize_cache;
-		break;
-	case UNMAP:
-		size = get_unaligned_be16(&cdb[7]);
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_unmap;
-		break;
-	case WRITE_SAME_16:
-		sectors = transport_get_sectors_16(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
-
-		if (sectors)
-			size = transport_get_size(1, cdb, cmd);
-		else {
-			pr_err("WSNZ=1, WRITE_SAME w/sectors=0 not supported\n");
-			goto out_invalid_cdb_field;
 		}
 
-		cmd->t_task_lba = get_unaligned_be64(&cdb[2]);
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-
-		if (target_check_write_same_discard(&cdb[1], dev) < 0)
-			goto out_unsupported_cdb;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_write_same;
-		break;
-	case WRITE_SAME:
-		sectors = transport_get_sectors_10(cdb, cmd, &sector_ret);
-		if (sector_ret)
-			goto out_unsupported_cdb;
+		cmd = list_entry(dev->delayed_cmd_list.next,
+				 struct se_cmd, se_delayed_node);
+		list_del(&cmd->se_delayed_node);
+		spin_unlock(&dev->delayed_cmd_lock);
 
-		if (sectors)
-			size = transport_get_size(1, cdb, cmd);
-		else {
-			pr_err("WSNZ=1, WRITE_SAME w/sectors=0 not supported\n");
-			goto out_invalid_cdb_field;
-		}
+		__target_execute_cmd(cmd);
 
-		cmd->t_task_lba = get_unaligned_be32(&cdb[2]);
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		/*
-		 * Follow sbcr26 with WRITE_SAME (10) and check for the existence
-		 * of byte 1 bit 3 UNMAP instead of original reserved field
-		 */
-		if (target_check_write_same_discard(&cdb[1], dev) < 0)
-			goto out_unsupported_cdb;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_write_same;
-		break;
-	case ALLOW_MEDIUM_REMOVAL:
-	case ERASE:
-	case REZERO_UNIT:
-	case SEEK_10:
-	case SPACE:
-	case START_STOP:
-	case TEST_UNIT_READY:
-	case VERIFY:
-	case WRITE_FILEMARKS:
-		cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
-		if (!passthrough)
-			cmd->execute_cmd = target_emulate_noop;
-		break;
-	case GPCMD_CLOSE_TRACK:
-	case INITIALIZE_ELEMENT_STATUS:
-	case GPCMD_LOAD_UNLOAD:
-	case GPCMD_SET_SPEED:
-	case MOVE_MEDIUM:
-		cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
-		break;
-	case REPORT_LUNS:
-		cmd->execute_cmd = target_report_luns;
-		size = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
-		/*
-		 * Do implict HEAD_OF_QUEUE processing for REPORT_LUNS
-		 * See spc4r17 section 5.3
-		 */
-		if (cmd->se_dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
-			cmd->sam_task_attr = MSG_HEAD_TAG;
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case GET_EVENT_STATUS_NOTIFICATION:
-		size = (cdb[7] << 8) | cdb[8];
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	case ATA_16:
-		/* Only support ATA passthrough to pSCSI backends.. */
-		if (!passthrough)
-			goto out_unsupported_cdb;
-
-		/* T_LENGTH */
-		switch (cdb[2] & 0x3) {
-		case 0x0:
-			sectors = 0;
-			break;
-		case 0x1:
-			sectors = (((cdb[1] & 0x1) ? cdb[3] : 0) << 8) | cdb[4];
-			break;
-		case 0x2:
-			sectors = (((cdb[1] & 0x1) ? cdb[5] : 0) << 8) | cdb[6];
+		if (cmd->sam_task_attr == MSG_ORDERED_TAG)
 			break;
-		case 0x3:
-			pr_err("T_LENGTH=0x3 not supported for ATA_16\n");
-			goto out_invalid_cdb_field;
-		}
-
-		/* BYTE_BLOCK */
-		if (cdb[2] & 0x4) {
-			/* BLOCK T_TYPE: 512 or sector */
-			size = sectors * ((cdb[2] & 0x10) ?
-				dev->se_sub_dev->se_dev_attrib.block_size : 512);
-		} else {
-			/* BYTE */
-			size = sectors;
-		}
-		cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
-		break;
-	default:
-		pr_warn("TARGET_CORE[%s]: Unsupported SCSI Opcode"
-			" 0x%02x, sending CHECK_CONDITION.\n",
-			cmd->se_tfo->get_fabric_name(), cdb[0]);
-		goto out_unsupported_cdb;
-	}
-
-	if (cmd->unknown_data_length)
-		cmd->data_length = size;
-
-	if (size != cmd->data_length) {
-		pr_warn("TARGET_CORE[%s]: Expected Transfer Length:"
-			" %u does not match SCSI CDB Length: %u for SAM Opcode:"
-			" 0x%02x\n", cmd->se_tfo->get_fabric_name(),
-				cmd->data_length, size, cdb[0]);
-
-		cmd->cmd_spdtl = size;
-
-		if (cmd->data_direction == DMA_TO_DEVICE) {
-			pr_err("Rejecting underflow/overflow"
-					" WRITE data\n");
-			goto out_invalid_cdb_field;
-		}
-		/*
-		 * Reject READ_* or WRITE_* with overflow/underflow for
-		 * type SCF_SCSI_DATA_SG_IO_CDB.
-		 */
-		if (!ret && (dev->se_sub_dev->se_dev_attrib.block_size != 512))  {
-			pr_err("Failing OVERFLOW/UNDERFLOW for LBA op"
-				" CDB on non 512-byte sector setup subsystem"
-				" plugin: %s\n", dev->transport->name);
-			/* Returns CHECK_CONDITION + INVALID_CDB_FIELD */
-			goto out_invalid_cdb_field;
-		}
-
-		if (size > cmd->data_length) {
-			cmd->se_cmd_flags |= SCF_OVERFLOW_BIT;
-			cmd->residual_count = (size - cmd->data_length);
-		} else {
-			cmd->se_cmd_flags |= SCF_UNDERFLOW_BIT;
-			cmd->residual_count = (cmd->data_length - size);
-		}
-		cmd->data_length = size;
 	}
-
-	if (cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) {
-		if (sectors > su_dev->se_dev_attrib.fabric_max_sectors) {
-			printk_ratelimited(KERN_ERR "SCSI OP %02xh with too"
-				" big sectors %u exceeds fabric_max_sectors:"
-				" %u\n", cdb[0], sectors,
-				su_dev->se_dev_attrib.fabric_max_sectors);
-			goto out_invalid_cdb_field;
-		}
-		if (sectors > su_dev->se_dev_attrib.hw_max_sectors) {
-			printk_ratelimited(KERN_ERR "SCSI OP %02xh with too"
-				" big sectors %u exceeds backend hw_max_sectors:"
-				" %u\n", cdb[0], sectors,
-				su_dev->se_dev_attrib.hw_max_sectors);
-			goto out_invalid_cdb_field;
-		}
-	}
-
-	/* reject any command that we don't have a handler for */
-	if (!(passthrough || cmd->execute_cmd ||
-	     (cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB)))
-		goto out_unsupported_cdb;
-
-	transport_set_supported_SAM_opcode(cmd);
-	return ret;
-
-out_unsupported_cdb:
-	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-	cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
-	return -EINVAL;
-out_invalid_cdb_field:
-	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-	cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
-	return -EINVAL;
 }
 
 /*
@@ -3052,8 +1906,6 @@ out_invalid_cdb_field:
 static void transport_complete_task_attr(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
-	struct se_cmd *cmd_p, *cmd_tmp;
-	int new_active_tasks = 0;
 
 	if (cmd->sam_task_attr == MSG_SIMPLE_TAG) {
 		atomic_dec(&dev->simple_cmds);
@@ -3075,38 +1927,8 @@ static void transport_complete_task_attr(struct se_cmd *cmd)
 		pr_debug("Incremented dev_cur_ordered_id: %u for ORDERED:"
 			" %u\n", dev->dev_cur_ordered_id, cmd->se_ordered_id);
 	}
-	/*
-	 * Process all commands up to the last received
-	 * ORDERED task attribute which requires another blocking
-	 * boundary
-	 */
-	spin_lock(&dev->delayed_cmd_lock);
-	list_for_each_entry_safe(cmd_p, cmd_tmp,
-			&dev->delayed_cmd_list, se_delayed_node) {
-
-		list_del(&cmd_p->se_delayed_node);
-		spin_unlock(&dev->delayed_cmd_lock);
-
-		pr_debug("Calling add_tasks() for"
-			" cmd_p: 0x%02x Task Attr: 0x%02x"
-			" Dormant -> Active, se_ordered_id: %u\n",
-			cmd_p->t_task_cdb[0],
-			cmd_p->sam_task_attr, cmd_p->se_ordered_id);
 
-		target_add_to_execute_list(cmd_p);
-		new_active_tasks++;
-
-		spin_lock(&dev->delayed_cmd_lock);
-		if (cmd_p->sam_task_attr == MSG_ORDERED_TAG)
-			break;
-	}
-	spin_unlock(&dev->delayed_cmd_lock);
-	/*
-	 * If new tasks have become active, wake up the transport thread
-	 * to do the processing of the Active tasks.
-	 */
-	if (new_active_tasks != 0)
-		wake_up_interruptible(&dev->dev_queue_obj.thread_wq);
+	target_restart_delayed_cmds(dev);
 }
 
 static void transport_complete_qf(struct se_cmd *cmd)
@@ -3365,31 +2187,27 @@ int transport_generic_map_mem_to_cmd(
 	if (!sgl || !sgl_count)
 		return 0;
 
-	if ((cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) ||
-	    (cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB)) {
-		/*
-		 * Reject SCSI data overflow with map_mem_to_cmd() as incoming
-		 * scatterlists already have been set to follow what the fabric
-		 * passes for the original expected data transfer length.
-		 */
-		if (cmd->se_cmd_flags & SCF_OVERFLOW_BIT) {
-			pr_warn("Rejecting SCSI DATA overflow for fabric using"
-				" SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC\n");
-			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-			cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
-			return -EINVAL;
-		}
+	/*
+	 * Reject SCSI data overflow with map_mem_to_cmd() as incoming
+	 * scatterlists already have been set to follow what the fabric
+	 * passes for the original expected data transfer length.
+	 */
+	if (cmd->se_cmd_flags & SCF_OVERFLOW_BIT) {
+		pr_warn("Rejecting SCSI DATA overflow for fabric using"
+			" SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC\n");
+		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
+	}
 
-		cmd->t_data_sg = sgl;
-		cmd->t_data_nents = sgl_count;
+	cmd->t_data_sg = sgl;
+	cmd->t_data_nents = sgl_count;
 
-		if (sgl_bidi && sgl_bidi_count) {
-			cmd->t_bidi_data_sg = sgl_bidi;
-			cmd->t_bidi_data_nents = sgl_bidi_count;
-		}
-		cmd->se_cmd_flags |= SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC;
+	if (sgl_bidi && sgl_bidi_count) {
+		cmd->t_bidi_data_sg = sgl_bidi;
+		cmd->t_bidi_data_nents = sgl_bidi_count;
 	}
-
+	cmd->se_cmd_flags |= SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC;
 	return 0;
 }
 EXPORT_SYMBOL(transport_generic_map_mem_to_cmd);
@@ -3461,7 +2279,7 @@ transport_generic_get_mem(struct se_cmd *cmd)
 	cmd->t_data_nents = nents;
 	sg_init_table(cmd->t_data_sg, nents);
 
-	zero_flag = cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB ? 0 : __GFP_ZERO;
+	zero_flag = cmd->se_cmd_flags & SCF_SCSI_DATA_CDB ? 0 : __GFP_ZERO;
 
 	while (length) {
 		u32 page_len = min_t(u32, length, PAGE_SIZE);
@@ -3492,7 +2310,6 @@ out:
  */
 int transport_generic_new_cmd(struct se_cmd *cmd)
 {
-	struct se_device *dev = cmd->se_dev;
 	int ret = 0;
 
 	/*
@@ -3508,8 +2325,7 @@ int transport_generic_new_cmd(struct se_cmd *cmd)
 	}
 
 	/* Workaround for handling zero-length control CDBs */
-	if ((cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB) &&
-	    !cmd->data_length) {
+	if (!(cmd->se_cmd_flags & SCF_SCSI_DATA_CDB) && !cmd->data_length) {
 		spin_lock_irq(&cmd->t_state_lock);
 		cmd->t_state = TRANSPORT_COMPLETE;
 		cmd->transport_state |= CMD_T_ACTIVE;
@@ -3527,52 +2343,45 @@ int transport_generic_new_cmd(struct se_cmd *cmd)
 		return 0;
 	}
 
-	if (cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) {
-		struct se_dev_attrib *attr = &dev->se_sub_dev->se_dev_attrib;
-
-		if (transport_cmd_get_valid_sectors(cmd) < 0)
-			return -EINVAL;
-
-		BUG_ON(cmd->data_length % attr->block_size);
-		BUG_ON(DIV_ROUND_UP(cmd->data_length, attr->block_size) >
-			attr->hw_max_sectors);
-	}
-
 	atomic_inc(&cmd->t_fe_count);
 
 	/*
-	 * For WRITEs, let the fabric know its buffer is ready.
-	 *
-	 * The command will be added to the execution queue after its write
-	 * data has arrived.
+	 * If this command is not a write we can execute it right here,
+	 * for write buffers we need to notify the fabric driver first
+	 * and let it call back once the write buffers are ready.
 	 */
-	if (cmd->data_direction == DMA_TO_DEVICE) {
-		target_add_to_state_list(cmd);
-		return transport_generic_write_pending(cmd);
+	target_add_to_state_list(cmd);
+	if (cmd->data_direction != DMA_TO_DEVICE) {
+		target_execute_cmd(cmd);
+		return 0;
 	}
-	/*
-	 * Everything else but a WRITE, add the command to the execution queue.
-	 */
-	transport_execute_tasks(cmd);
-	return 0;
+
+	spin_lock_irq(&cmd->t_state_lock);
+	cmd->t_state = TRANSPORT_WRITE_PENDING;
+	spin_unlock_irq(&cmd->t_state_lock);
+
+	transport_cmd_check_stop(cmd, false);
+
+	ret = cmd->se_tfo->write_pending(cmd);
+	if (ret == -EAGAIN || ret == -ENOMEM)
+		goto queue_full;
+
+	if (ret < 0)
+		return ret;
+	return 1;
 
 out_fail:
 	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
 	cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 	return -EINVAL;
+queue_full:
+	pr_debug("Handling write_pending QUEUE__FULL: se_cmd: %p\n", cmd);
+	cmd->t_state = TRANSPORT_COMPLETE_QF_WP;
+	transport_handle_queue_full(cmd, cmd->se_dev);
+	return 0;
 }
 EXPORT_SYMBOL(transport_generic_new_cmd);
 
-/*	transport_generic_process_write():
- *
- *
- */
-void transport_generic_process_write(struct se_cmd *cmd)
-{
-	transport_execute_tasks(cmd);
-}
-EXPORT_SYMBOL(transport_generic_process_write);
-
 static void transport_write_pending_qf(struct se_cmd *cmd)
 {
 	int ret;
@@ -3585,43 +2394,6 @@ static void transport_write_pending_qf(struct se_cmd *cmd)
 	}
 }
 
-static int transport_generic_write_pending(struct se_cmd *cmd)
-{
-	unsigned long flags;
-	int ret;
-
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
-	cmd->t_state = TRANSPORT_WRITE_PENDING;
-	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-
-	/*
-	 * Clear the se_cmd for WRITE_PENDING status in order to set
-	 * CMD_T_ACTIVE so that transport_generic_handle_data can be called
-	 * from HW target mode interrupt code.  This is safe to be called
-	 * with transport_off=1 before the cmd->se_tfo->write_pending
-	 * because the se_cmd->se_lun pointer is not being cleared.
-	 */
-	transport_cmd_check_stop(cmd, 1, 0);
-
-	/*
-	 * Call the fabric write_pending function here to let the
-	 * frontend know that WRITE buffers are ready.
-	 */
-	ret = cmd->se_tfo->write_pending(cmd);
-	if (ret == -EAGAIN || ret == -ENOMEM)
-		goto queue_full;
-	else if (ret < 0)
-		return ret;
-
-	return 1;
-
-queue_full:
-	pr_debug("Handling write_pending QUEUE__FULL: se_cmd: %p\n", cmd);
-	cmd->t_state = TRANSPORT_COMPLETE_QF_WP;
-	transport_handle_queue_full(cmd, cmd->se_dev);
-	return 0;
-}
-
 void transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks)
 {
 	if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD)) {
@@ -3648,10 +2420,11 @@ EXPORT_SYMBOL(transport_generic_free_cmd);
  * @se_cmd:	command descriptor to add
  * @ack_kref:	Signal that fabric will perform an ack target_put_sess_cmd()
  */
-void target_get_sess_cmd(struct se_session *se_sess, struct se_cmd *se_cmd,
-			bool ack_kref)
+static int target_get_sess_cmd(struct se_session *se_sess, struct se_cmd *se_cmd,
+			       bool ack_kref)
 {
 	unsigned long flags;
+	int ret = 0;
 
 	kref_init(&se_cmd->cmd_kref);
 	/*
@@ -3665,11 +2438,17 @@ void target_get_sess_cmd(struct se_session *se_sess, struct se_cmd *se_cmd,
 	}
 
 	spin_lock_irqsave(&se_sess->sess_cmd_lock, flags);
+	if (se_sess->sess_tearing_down) {
+		ret = -ESHUTDOWN;
+		goto out;
+	}
 	list_add_tail(&se_cmd->se_cmd_list, &se_sess->sess_cmd_list);
 	se_cmd->check_release = 1;
+
+out:
 	spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
+	return ret;
 }
-EXPORT_SYMBOL(target_get_sess_cmd);
 
 static void target_release_cmd_kref(struct kref *kref)
 {
@@ -3704,28 +2483,27 @@ int target_put_sess_cmd(struct se_session *se_sess, struct se_cmd *se_cmd)
 }
 EXPORT_SYMBOL(target_put_sess_cmd);
 
-/* target_splice_sess_cmd_list - Split active cmds into sess_wait_list
- * @se_sess:	session to split
+/* target_sess_cmd_list_set_waiting - Flag all commands in
+ *         sess_cmd_list to complete cmd_wait_comp.  Set
+ *         sess_tearing_down so no more commands are queued.
+ * @se_sess:	session to flag
  */
-void target_splice_sess_cmd_list(struct se_session *se_sess)
+void target_sess_cmd_list_set_waiting(struct se_session *se_sess)
 {
 	struct se_cmd *se_cmd;
 	unsigned long flags;
 
-	WARN_ON(!list_empty(&se_sess->sess_wait_list));
-	INIT_LIST_HEAD(&se_sess->sess_wait_list);
-
 	spin_lock_irqsave(&se_sess->sess_cmd_lock, flags);
-	se_sess->sess_tearing_down = 1;
 
-	list_splice_init(&se_sess->sess_cmd_list, &se_sess->sess_wait_list);
+	WARN_ON(se_sess->sess_tearing_down);
+	se_sess->sess_tearing_down = 1;
 
-	list_for_each_entry(se_cmd, &se_sess->sess_wait_list, se_cmd_list)
+	list_for_each_entry(se_cmd, &se_sess->sess_cmd_list, se_cmd_list)
 		se_cmd->cmd_wait_set = 1;
 
 	spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
 }
-EXPORT_SYMBOL(target_splice_sess_cmd_list);
+EXPORT_SYMBOL(target_sess_cmd_list_set_waiting);
 
 /* target_wait_for_sess_cmds - Wait for outstanding descriptors
  * @se_sess:    session to wait for active I/O
@@ -3739,7 +2517,7 @@ void target_wait_for_sess_cmds(
 	bool rc = false;
 
 	list_for_each_entry_safe(se_cmd, tmp_cmd,
-				&se_sess->sess_wait_list, se_cmd_list) {
+				&se_sess->sess_cmd_list, se_cmd_list) {
 		list_del(&se_cmd->se_cmd_list);
 
 		pr_debug("Waiting for se_cmd: %p t_state: %d, fabric state:"
@@ -3791,26 +2569,20 @@ static int transport_lun_wait_for_tasks(struct se_cmd *cmd, struct se_lun *lun)
 		pr_debug("ConfigFS ITT[0x%08x] - CMD_T_STOP, skipping\n",
 			 cmd->se_tfo->get_task_tag(cmd));
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-		transport_cmd_check_stop(cmd, 1, 0);
+		transport_cmd_check_stop(cmd, false);
 		return -EPERM;
 	}
 	cmd->transport_state |= CMD_T_LUN_FE_STOP;
 	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
-	wake_up_interruptible(&cmd->se_dev->dev_queue_obj.thread_wq);
-
 	// XXX: audit task_flags checks.
 	spin_lock_irqsave(&cmd->t_state_lock, flags);
 	if ((cmd->transport_state & CMD_T_BUSY) &&
 	    (cmd->transport_state & CMD_T_SENT)) {
 		if (!target_stop_cmd(cmd, &flags))
 			ret++;
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-	} else {
-		spin_unlock_irqrestore(&cmd->t_state_lock,
-				flags);
-		target_remove_from_execute_list(cmd);
 	}
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
 	pr_debug("ConfigFS: cmd: %p stop tasks ret:"
 			" %d\n", cmd, ret);
@@ -3821,7 +2593,6 @@ static int transport_lun_wait_for_tasks(struct se_cmd *cmd, struct se_lun *lun)
 		pr_debug("ConfigFS: ITT[0x%08x] - stopped cmd....\n",
 				cmd->se_tfo->get_task_tag(cmd));
 	}
-	transport_remove_cmd_from_queue(cmd);
 
 	return 0;
 }
@@ -3840,11 +2611,6 @@ static void __transport_clear_lun_from_sessions(struct se_lun *lun)
 		       struct se_cmd, se_lun_node);
 		list_del_init(&cmd->se_lun_node);
 
-		/*
-		 * This will notify iscsi_target_transport.c:
-		 * transport_cmd_check_stop() that a LUN shutdown is in
-		 * progress for the iscsi_cmd_t.
-		 */
 		spin_lock(&cmd->t_state_lock);
 		pr_debug("SE_LUN[%d] - Setting cmd->transport"
 			"_lun_stop for  ITT: 0x%08x\n",
@@ -3911,7 +2677,7 @@ check_cond:
 
 			spin_unlock_irqrestore(&cmd->t_state_lock,
 					cmd_flags);
-			transport_cmd_check_stop(cmd, 1, 0);
+			transport_cmd_check_stop(cmd, false);
 			complete(&cmd->transport_lun_fe_stop_comp);
 			spin_lock_irqsave(&lun->lun_cmd_lock, lun_flags);
 			continue;
@@ -3967,10 +2733,7 @@ bool transport_wait_for_tasks(struct se_cmd *cmd)
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 		return false;
 	}
-	/*
-	 * Only perform a possible wait_for_tasks if SCF_SUPPORTED_SAM_OPCODE
-	 * has been set in transport_set_supported_SAM_opcode().
-	 */
+
 	if (!(cmd->se_cmd_flags & SCF_SUPPORTED_SAM_OPCODE) &&
 	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB)) {
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
@@ -4028,8 +2791,6 @@ bool transport_wait_for_tasks(struct se_cmd *cmd)
 
 	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
-	wake_up_interruptible(&cmd->se_dev->dev_queue_obj.thread_wq);
-
 	wait_for_completion(&cmd->t_transport_stop_comp);
 
 	spin_lock_irqsave(&cmd->t_state_lock, flags);
@@ -4212,6 +2973,15 @@ int transport_send_check_condition_and_sense(
 		/* WRITE PROTECTED */
 		buffer[offset+SPC_ASC_KEY_OFFSET] = 0x27;
 		break;
+	case TCM_ADDRESS_OUT_OF_RANGE:
+		/* CURRENT ERROR */
+		buffer[offset] = 0x70;
+		buffer[offset+SPC_ADD_SENSE_LEN_OFFSET] = 10;
+		/* ILLEGAL REQUEST */
+		buffer[offset+SPC_SENSE_KEY_OFFSET] = ILLEGAL_REQUEST;
+		/* LOGICAL BLOCK ADDRESS OUT OF RANGE */
+		buffer[offset+SPC_ASC_KEY_OFFSET] = 0x21;
+		break;
 	case TCM_CHECK_CONDITION_UNIT_ATTENTION:
 		/* CURRENT ERROR */
 		buffer[offset] = 0x70;
@@ -4312,8 +3082,9 @@ void transport_send_task_abort(struct se_cmd *cmd)
 	cmd->se_tfo->queue_status(cmd);
 }
 
-static int transport_generic_do_tmr(struct se_cmd *cmd)
+static void target_tmr_work(struct work_struct *work)
 {
+	struct se_cmd *cmd = container_of(work, struct se_cmd, work);
 	struct se_device *dev = cmd->se_dev;
 	struct se_tmr_req *tmr = cmd->se_tmr_req;
 	int ret;
@@ -4349,80 +3120,13 @@ static int transport_generic_do_tmr(struct se_cmd *cmd)
 	cmd->se_tfo->queue_tm_rsp(cmd);
 
 	transport_cmd_check_stop_to_fabric(cmd);
-	return 0;
 }
 
-/*	transport_processing_thread():
- *
- *
- */
-static int transport_processing_thread(void *param)
+int transport_generic_handle_tmr(
+	struct se_cmd *cmd)
 {
-	int ret;
-	struct se_cmd *cmd;
-	struct se_device *dev = param;
-
-	while (!kthread_should_stop()) {
-		ret = wait_event_interruptible(dev->dev_queue_obj.thread_wq,
-				atomic_read(&dev->dev_queue_obj.queue_cnt) ||
-				kthread_should_stop());
-		if (ret < 0)
-			goto out;
-
-get_cmd:
-		cmd = transport_get_cmd_from_queue(&dev->dev_queue_obj);
-		if (!cmd)
-			continue;
-
-		switch (cmd->t_state) {
-		case TRANSPORT_NEW_CMD:
-			BUG();
-			break;
-		case TRANSPORT_NEW_CMD_MAP:
-			if (!cmd->se_tfo->new_cmd_map) {
-				pr_err("cmd->se_tfo->new_cmd_map is"
-					" NULL for TRANSPORT_NEW_CMD_MAP\n");
-				BUG();
-			}
-			ret = cmd->se_tfo->new_cmd_map(cmd);
-			if (ret < 0) {
-				transport_generic_request_failure(cmd);
-				break;
-			}
-			ret = transport_generic_new_cmd(cmd);
-			if (ret < 0) {
-				transport_generic_request_failure(cmd);
-				break;
-			}
-			break;
-		case TRANSPORT_PROCESS_WRITE:
-			transport_generic_process_write(cmd);
-			break;
-		case TRANSPORT_PROCESS_TMR:
-			transport_generic_do_tmr(cmd);
-			break;
-		case TRANSPORT_COMPLETE_QF_WP:
-			transport_write_pending_qf(cmd);
-			break;
-		case TRANSPORT_COMPLETE_QF_OK:
-			transport_complete_qf(cmd);
-			break;
-		default:
-			pr_err("Unknown t_state: %d  for ITT: 0x%08x "
-				"i_state: %d on SE LUN: %u\n",
-				cmd->t_state,
-				cmd->se_tfo->get_task_tag(cmd),
-				cmd->se_tfo->get_cmd_state(cmd),
-				cmd->se_lun->unpacked_lun);
-			BUG();
-		}
-
-		goto get_cmd;
-	}
-
-out:
-	WARN_ON(!list_empty(&dev->state_list));
-	WARN_ON(!list_empty(&dev->dev_queue_obj.qobj_list));
-	dev->process_thread = NULL;
+	INIT_WORK(&cmd->work, target_tmr_work);
+	queue_work(cmd->se_dev->tmr_wq, &cmd->work);
 	return 0;
 }
+EXPORT_SYMBOL(transport_generic_handle_tmr);
diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c
index 5b65f33939a8..b9cb5006177e 100644
--- a/drivers/target/tcm_fc/tfc_cmd.c
+++ b/drivers/target/tcm_fc/tfc_cmd.c
@@ -215,7 +215,7 @@ int ft_write_pending(struct se_cmd *se_cmd)
 		 */
 		if ((ep->xid <= lport->lro_xid) &&
 		    (fh->fh_r_ctl == FC_RCTL_DD_DATA_DESC)) {
-			if ((se_cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) &&
+			if ((se_cmd->se_cmd_flags & SCF_SCSI_DATA_CDB) &&
 			    lport->tt.ddp_target(lport, ep->xid,
 						 se_cmd->t_data_sg,
 						 se_cmd->t_data_nents))
@@ -543,9 +543,11 @@ static void ft_send_work(struct work_struct *work)
 	 * Use a single se_cmd->cmd_kref as we expect to release se_cmd
 	 * directly from ft_check_stop_free callback in response path.
 	 */
-	target_submit_cmd(&cmd->se_cmd, cmd->sess->se_sess, fcp->fc_cdb,
-			&cmd->ft_sense_buffer[0], scsilun_to_int(&fcp->fc_lun),
-			ntohl(fcp->fc_dl), task_attr, data_dir, 0);
+	if (target_submit_cmd(&cmd->se_cmd, cmd->sess->se_sess, fcp->fc_cdb,
+			      &cmd->ft_sense_buffer[0], scsilun_to_int(&fcp->fc_lun),
+			      ntohl(fcp->fc_dl), task_attr, data_dir, 0))
+		goto err;
+
 	pr_debug("r_ctl %x alloc target_submit_cmd\n", fh->fh_r_ctl);
 	return;
 
diff --git a/drivers/target/tcm_fc/tfc_io.c b/drivers/target/tcm_fc/tfc_io.c
index 071a505f98fc..ad36ede1a1ea 100644
--- a/drivers/target/tcm_fc/tfc_io.c
+++ b/drivers/target/tcm_fc/tfc_io.c
@@ -183,6 +183,13 @@ int ft_queue_data_in(struct se_cmd *se_cmd)
 	return ft_queue_status(se_cmd);
 }
 
+static void ft_execute_work(struct work_struct *work)
+{
+	struct ft_cmd *cmd = container_of(work, struct ft_cmd, work);
+
+	target_execute_cmd(&cmd->se_cmd);
+}
+
 /*
  * Receive write data frame.
  */
@@ -307,8 +314,10 @@ void ft_recv_write_data(struct ft_cmd *cmd, struct fc_frame *fp)
 		cmd->write_data_len += tlen;
 	}
 last_frame:
-	if (cmd->write_data_len == se_cmd->data_length)
-		transport_generic_handle_data(se_cmd);
+	if (cmd->write_data_len == se_cmd->data_length) {
+		INIT_WORK(&cmd->work, ft_execute_work);
+		queue_work(cmd->sess->tport->tpg->workqueue, &cmd->work);
+	}
 drop:
 	fc_frame_free(fp);
 }
diff --git a/drivers/usb/gadget/tcm_usb_gadget.c b/drivers/usb/gadget/tcm_usb_gadget.c
index c46439c8dd74..5444866e13ef 100644
--- a/drivers/usb/gadget/tcm_usb_gadget.c
+++ b/drivers/usb/gadget/tcm_usb_gadget.c
@@ -294,7 +294,7 @@ static int bot_send_write_request(struct usbg_cmd *cmd)
 		pr_err("%s(%d)\n", __func__, __LINE__);
 
 	wait_for_completion(&cmd->write_complete);
-	transport_generic_process_write(se_cmd);
+	target_execute_cmd(se_cmd);
 cleanup:
 	return ret;
 }
@@ -725,7 +725,7 @@ static int uasp_send_write_request(struct usbg_cmd *cmd)
 	}
 
 	wait_for_completion(&cmd->write_complete);
-	transport_generic_process_write(se_cmd);
+	target_execute_cmd(se_cmd);
 cleanup:
 	return ret;
 }
@@ -1065,16 +1065,20 @@ static void usbg_cmd_work(struct work_struct *work)
 				tv_nexus->tvn_se_sess->se_tpg->se_tpg_tfo,
 				tv_nexus->tvn_se_sess, cmd->data_len, DMA_NONE,
 				cmd->prio_attr, cmd->sense_iu.sense);
-
-		transport_send_check_condition_and_sense(se_cmd,
-				TCM_UNSUPPORTED_SCSI_OPCODE, 1);
-		usbg_cleanup_cmd(cmd);
-		return;
+		goto out;
 	}
 
-	target_submit_cmd(se_cmd, tv_nexus->tvn_se_sess,
+	if (target_submit_cmd(se_cmd, tv_nexus->tvn_se_sess,
 			cmd->cmd_buf, cmd->sense_iu.sense, cmd->unpacked_lun,
-			0, cmd->prio_attr, dir, TARGET_SCF_UNKNOWN_SIZE);
+			0, cmd->prio_attr, dir, TARGET_SCF_UNKNOWN_SIZE) < 0)
+		goto out;
+
+	return;
+
+out:
+	transport_send_check_condition_and_sense(se_cmd,
+			TCM_UNSUPPORTED_SCSI_OPCODE, 1);
+	usbg_cleanup_cmd(cmd);
 }
 
 static int usbg_submit_command(struct f_uas *fu,
@@ -1177,16 +1181,20 @@ static void bot_cmd_work(struct work_struct *work)
 				tv_nexus->tvn_se_sess->se_tpg->se_tpg_tfo,
 				tv_nexus->tvn_se_sess, cmd->data_len, DMA_NONE,
 				cmd->prio_attr, cmd->sense_iu.sense);
-
-		transport_send_check_condition_and_sense(se_cmd,
-				TCM_UNSUPPORTED_SCSI_OPCODE, 1);
-		usbg_cleanup_cmd(cmd);
-		return;
+		goto out;
 	}
 
-	target_submit_cmd(se_cmd, tv_nexus->tvn_se_sess,
+	if (target_submit_cmd(se_cmd, tv_nexus->tvn_se_sess,
 			cmd->cmd_buf, cmd->sense_iu.sense, cmd->unpacked_lun,
-			cmd->data_len, cmd->prio_attr, dir, 0);
+			cmd->data_len, cmd->prio_attr, dir, 0) < 0)
+		goto out;
+
+	return;
+
+out:
+	transport_send_check_condition_and_sense(se_cmd,
+				TCM_UNSUPPORTED_SCSI_OPCODE, 1);
+	usbg_cleanup_cmd(cmd);
 }
 
 static int bot_submit_command(struct f_uas *fu,
@@ -1400,19 +1408,6 @@ static u32 usbg_tpg_get_inst_index(struct se_portal_group *se_tpg)
 	return 1;
 }
 
-static int usbg_new_cmd(struct se_cmd *se_cmd)
-{
-	struct usbg_cmd *cmd = container_of(se_cmd, struct usbg_cmd,
-			se_cmd);
-	int ret;
-
-	ret = target_setup_cmd_from_cdb(se_cmd, cmd->cmd_buf);
-	if (ret)
-		return ret;
-
-	return transport_generic_map_mem_to_cmd(se_cmd, NULL, 0, NULL, 0);
-}
-
 static void usbg_cmd_release(struct kref *ref)
 {
 	struct usbg_cmd *cmd = container_of(ref, struct usbg_cmd,
@@ -1902,7 +1897,6 @@ static struct target_core_fabric_ops usbg_ops = {
 	.tpg_alloc_fabric_acl		= usbg_alloc_fabric_acl,
 	.tpg_release_fabric_acl		= usbg_release_fabric_acl,
 	.tpg_get_inst_index		= usbg_tpg_get_inst_index,
-	.new_cmd_map			= usbg_new_cmd,
 	.release_cmd			= usbg_release_cmd,
 	.shutdown_session		= usbg_shutdown_session,
 	.close_session			= usbg_close_session,
diff --git a/include/target/target_core_backend.h b/include/target/target_core_backend.h
index 2d7db85e93ae..f1405d335a96 100644
--- a/include/target/target_core_backend.h
+++ b/include/target/target_core_backend.h
@@ -24,10 +24,8 @@ struct se_subsystem_api {
 				struct se_subsystem_dev *, void *);
 	void (*free_device)(void *);
 	int (*transport_complete)(struct se_cmd *cmd, struct scatterlist *);
-	int (*execute_cmd)(struct se_cmd *, struct scatterlist *, u32,
-			enum dma_data_direction);
-	int (*do_discard)(struct se_device *, sector_t, u32);
-	void (*do_sync_cache)(struct se_cmd *);
+
+	int (*parse_cdb)(struct se_cmd *cmd);
 	ssize_t (*check_configfs_dev_params)(struct se_hba *,
 			struct se_subsystem_dev *);
 	ssize_t (*set_configfs_dev_params)(struct se_hba *,
@@ -40,6 +38,13 @@ struct se_subsystem_api {
 	unsigned char *(*get_sense_buffer)(struct se_cmd *);
 };
 
+struct spc_ops {
+	int (*execute_rw)(struct se_cmd *cmd);
+	int (*execute_sync_cache)(struct se_cmd *cmd);
+	int (*execute_write_same)(struct se_cmd *cmd);
+	int (*execute_unmap)(struct se_cmd *cmd);
+};
+
 int	transport_subsystem_register(struct se_subsystem_api *);
 void	transport_subsystem_release(struct se_subsystem_api *);
 
@@ -49,6 +54,10 @@ struct se_device *transport_add_device_to_core_hba(struct se_hba *,
 
 void	target_complete_cmd(struct se_cmd *, u8);
 
+int	sbc_parse_cdb(struct se_cmd *cmd, struct spc_ops *ops);
+int	spc_parse_cdb(struct se_cmd *cmd, unsigned int *size);
+int	spc_get_write_same_sectors(struct se_cmd *cmd);
+
 void	transport_set_vpd_proto_id(struct t10_vpd *, unsigned char *);
 int	transport_set_vpd_assoc(struct t10_vpd *, unsigned char *);
 int	transport_set_vpd_ident_type(struct t10_vpd *, unsigned char *);
diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h
index dc35d8660aa6..128ce46fa48a 100644
--- a/include/target/target_core_base.h
+++ b/include/target/target_core_base.h
@@ -145,12 +145,9 @@ enum transport_state_table {
 	TRANSPORT_NO_STATE	= 0,
 	TRANSPORT_NEW_CMD	= 1,
 	TRANSPORT_WRITE_PENDING	= 3,
-	TRANSPORT_PROCESS_WRITE	= 4,
 	TRANSPORT_PROCESSING	= 5,
 	TRANSPORT_COMPLETE	= 6,
-	TRANSPORT_PROCESS_TMR	= 9,
 	TRANSPORT_ISTATE_PROCESSING = 11,
-	TRANSPORT_NEW_CMD_MAP	= 16,
 	TRANSPORT_COMPLETE_QF_WP = 18,
 	TRANSPORT_COMPLETE_QF_OK = 19,
 };
@@ -160,25 +157,20 @@ enum se_cmd_flags_table {
 	SCF_SUPPORTED_SAM_OPCODE	= 0x00000001,
 	SCF_TRANSPORT_TASK_SENSE	= 0x00000002,
 	SCF_EMULATED_TASK_SENSE		= 0x00000004,
-	SCF_SCSI_DATA_SG_IO_CDB		= 0x00000008,
-	SCF_SCSI_CONTROL_SG_IO_CDB	= 0x00000010,
-	SCF_SCSI_NON_DATA_CDB		= 0x00000020,
-	SCF_SCSI_TMR_CDB		= 0x00000040,
-	SCF_SCSI_CDB_EXCEPTION		= 0x00000080,
-	SCF_SCSI_RESERVATION_CONFLICT	= 0x00000100,
-	SCF_FUA				= 0x00000200,
-	SCF_SE_LUN_CMD			= 0x00000800,
-	SCF_SE_ALLOW_EOO		= 0x00001000,
-	SCF_BIDI			= 0x00002000,
-	SCF_SENT_CHECK_CONDITION	= 0x00004000,
-	SCF_OVERFLOW_BIT		= 0x00008000,
-	SCF_UNDERFLOW_BIT		= 0x00010000,
-	SCF_SENT_DELAYED_TAS		= 0x00020000,
-	SCF_ALUA_NON_OPTIMIZED		= 0x00040000,
-	SCF_DELAYED_CMD_FROM_SAM_ATTR	= 0x00080000,
-	SCF_UNUSED			= 0x00100000,
-	SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC = 0x00200000,
-	SCF_ACK_KREF			= 0x00400000,
+	SCF_SCSI_DATA_CDB		= 0x00000008,
+	SCF_SCSI_TMR_CDB		= 0x00000010,
+	SCF_SCSI_CDB_EXCEPTION		= 0x00000020,
+	SCF_SCSI_RESERVATION_CONFLICT	= 0x00000040,
+	SCF_FUA				= 0x00000080,
+	SCF_SE_LUN_CMD			= 0x00000100,
+	SCF_BIDI			= 0x00000400,
+	SCF_SENT_CHECK_CONDITION	= 0x00000800,
+	SCF_OVERFLOW_BIT		= 0x00001000,
+	SCF_UNDERFLOW_BIT		= 0x00002000,
+	SCF_SENT_DELAYED_TAS		= 0x00004000,
+	SCF_ALUA_NON_OPTIMIZED		= 0x00008000,
+	SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC = 0x00020000,
+	SCF_ACK_KREF			= 0x00040000,
 };
 
 /* struct se_dev_entry->lun_flags and struct se_lun->lun_access */
@@ -220,6 +212,7 @@ enum tcm_sense_reason_table {
 	TCM_CHECK_CONDITION_UNIT_ATTENTION	= 0x0e,
 	TCM_CHECK_CONDITION_NOT_READY		= 0x0f,
 	TCM_RESERVATION_CONFLICT		= 0x10,
+	TCM_ADDRESS_OUT_OF_RANGE		= 0x11,
 };
 
 enum target_sc_flags_table {
@@ -471,13 +464,6 @@ struct t10_reservation {
 	struct t10_reservation_ops pr_ops;
 };
 
-struct se_queue_obj {
-	atomic_t		queue_cnt;
-	spinlock_t		cmd_queue_lock;
-	struct list_head	qobj_list;
-	wait_queue_head_t	thread_wq;
-};
-
 struct se_tmr_req {
 	/* Task Management function to be performed */
 	u8			function;
@@ -486,11 +472,8 @@ struct se_tmr_req {
 	int			call_transport;
 	/* Reference to ITT that Task Mgmt should be performed */
 	u32			ref_task_tag;
-	/* 64-bit encoded SAM LUN from $FABRIC_MOD TMR header */
-	u64			ref_task_lun;
 	void 			*fabric_tmr_ptr;
 	struct se_cmd		*task_cmd;
-	struct se_cmd		*ref_cmd;
 	struct se_device	*tmr_dev;
 	struct se_lun		*tmr_lun;
 	struct list_head	tmr_list;
@@ -537,7 +520,6 @@ struct se_cmd {
 	/* Only used for internal passthrough and legacy TCM fabric modules */
 	struct se_session	*se_sess;
 	struct se_tmr_req	*se_tmr_req;
-	struct list_head	se_queue_node;
 	struct list_head	se_cmd_list;
 	struct completion	cmd_wait_comp;
 	struct kref		cmd_kref;
@@ -575,7 +557,6 @@ struct se_cmd {
 	struct scatterlist	*t_bidi_data_sg;
 	unsigned int		t_bidi_data_nents;
 
-	struct list_head	execute_list;
 	struct list_head	state_list;
 	bool			state_active;
 
@@ -633,7 +614,6 @@ struct se_session {
 	struct list_head	sess_list;
 	struct list_head	sess_acl_list;
 	struct list_head	sess_cmd_list;
-	struct list_head	sess_wait_list;
 	spinlock_t		sess_cmd_lock;
 	struct kref		sess_kref;
 };
@@ -780,13 +760,11 @@ struct se_device {
 	/* Active commands on this virtual SE device */
 	atomic_t		simple_cmds;
 	atomic_t		dev_ordered_id;
-	atomic_t		execute_tasks;
 	atomic_t		dev_ordered_sync;
 	atomic_t		dev_qf_count;
 	struct se_obj		dev_obj;
 	struct se_obj		dev_access_obj;
 	struct se_obj		dev_export_obj;
-	struct se_queue_obj	dev_queue_obj;
 	spinlock_t		delayed_cmd_lock;
 	spinlock_t		execute_task_lock;
 	spinlock_t		dev_reservation_lock;
@@ -802,11 +780,9 @@ struct se_device {
 	struct t10_pr_registration *dev_pr_res_holder;
 	struct list_head	dev_sep_list;
 	struct list_head	dev_tmr_list;
-	/* Pointer to descriptor for processing thread */
-	struct task_struct	*process_thread;
+	struct workqueue_struct *tmr_wq;
 	struct work_struct	qf_work_queue;
 	struct list_head	delayed_cmd_list;
-	struct list_head	execute_list;
 	struct list_head	state_list;
 	struct list_head	qf_cmd_list;
 	/* Pointer to associated SE HBA */
diff --git a/include/target/target_core_fabric.h b/include/target/target_core_fabric.h
index c78a23333c4f..69fb3cfd02d7 100644
--- a/include/target/target_core_fabric.h
+++ b/include/target/target_core_fabric.h
@@ -33,12 +33,6 @@ struct target_core_fabric_ops {
 					struct se_node_acl *);
 	u32 (*tpg_get_inst_index)(struct se_portal_group *);
 	/*
-	 * Optional function pointer for TCM to perform command map
-	 * from TCM processing thread context, for those struct se_cmd
-	 * initially allocated in interrupt context.
-	 */
-	int (*new_cmd_map)(struct se_cmd *);
-	/*
 	 * Optional to release struct se_cmd and fabric dependent allocated
 	 * I/O descriptor in transport_cmd_check_stop().
 	 *
@@ -108,20 +102,18 @@ void	transport_init_se_cmd(struct se_cmd *, struct target_core_fabric_ops *,
 		struct se_session *, u32, int, int, unsigned char *);
 int	transport_lookup_cmd_lun(struct se_cmd *, u32);
 int	target_setup_cmd_from_cdb(struct se_cmd *, unsigned char *);
-void	target_submit_cmd(struct se_cmd *, struct se_session *, unsigned char *,
+int	target_submit_cmd(struct se_cmd *, struct se_session *, unsigned char *,
 		unsigned char *, u32, u32, int, int, int);
 int	target_submit_tmr(struct se_cmd *se_cmd, struct se_session *se_sess,
 		unsigned char *sense, u32 unpacked_lun,
 		void *fabric_tmr_ptr, unsigned char tm_type,
 		gfp_t, unsigned int, int);
 int	transport_handle_cdb_direct(struct se_cmd *);
-int	transport_generic_handle_cdb_map(struct se_cmd *);
-int	transport_generic_handle_data(struct se_cmd *);
 int	transport_generic_map_mem_to_cmd(struct se_cmd *cmd,
 		struct scatterlist *, u32, struct scatterlist *, u32);
 int	transport_generic_new_cmd(struct se_cmd *);
 
-void	transport_generic_process_write(struct se_cmd *);
+void	target_execute_cmd(struct se_cmd *cmd);
 
 void	transport_generic_free_cmd(struct se_cmd *, int);
 
@@ -129,9 +121,8 @@ bool	transport_wait_for_tasks(struct se_cmd *);
 int	transport_check_aborted_status(struct se_cmd *, int);
 int	transport_send_check_condition_and_sense(struct se_cmd *, u8, int);
 
-void	target_get_sess_cmd(struct se_session *, struct se_cmd *, bool);
 int	target_put_sess_cmd(struct se_session *, struct se_cmd *);
-void	target_splice_sess_cmd_list(struct se_session *);
+void	target_sess_cmd_list_set_waiting(struct se_session *);
 void	target_wait_for_sess_cmds(struct se_session *, int);
 
 int	core_alua_check_nonop_delay(struct se_cmd *);