target: Fix remote-port TMR ABORT + se_cmd fabric stop

commit 0f4a9431 upstream.

To address the bug where fabric driver level shutdown
of se_cmd occurs at the same time when TMR CMD_T_ABORTED
is happening resulting in a -1 ->cmd_kref, this patch
adds a CMD_T_FABRIC_STOP bit that is used to determine
when TMR + driver I_T nexus shutdown is happening
concurrently.

It changes target_sess_cmd_list_set_waiting() to obtain
se_cmd->cmd_kref + set CMD_T_FABRIC_STOP, and drop local
reference in target_wait_for_sess_cmds() and invoke extra
target_put_sess_cmd() during Task Aborted Status (TAS)
when necessary.

Also, it adds a new target_wait_free_cmd() wrapper around
transport_wait_for_tasks() for the special case within
transport_generic_free_cmd() to set CMD_T_FABRIC_STOP,
and is now aware of CMD_T_ABORTED + CMD_T_TAS status
bits to know when an extra transport_put_cmd() during
TAS is required.

Note transport_generic_free_cmd() is expected to block on
cmd->cmd_wait_comp in order to follow what iscsi-target
expects during iscsi_conn context se_cmd shutdown.

Cc: Quinn Tran <quinn.tran@qlogic.com>
Cc: Himanshu Madhani <himanshu.madhani@qlogic.com>
Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Andy Grover <agrover@redhat.com>
Cc: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@daterainc.com>
[ luis: backported to 3.16: used Nicholas' backport to 3.14 ]
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>

drivers/target/target_core_tmr.c

@@ -82,16 +82,18 @@ void core_tmr_release_req(
 	kfree(tmr);
 }
 
-static void core_tmr_handle_tas_abort(
-	struct se_session *tmr_sess,
-	struct se_cmd *cmd,
-	int tas)
+static void core_tmr_handle_tas_abort(struct se_cmd *cmd, int tas)
 {
-	bool remove = true;
+	unsigned long flags;
+	bool remove = true, send_tas;
 	/*
 	 * TASK ABORTED status (TAS) bit support
-	*/
-	if (tmr_sess && tmr_sess != cmd->se_sess && tas) {
+	 */
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	send_tas = (cmd->transport_state & CMD_T_TAS);
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+
+	if (send_tas) {
 		remove = false;
 		transport_send_task_abort(cmd);
 	}
@@ -114,7 +116,8 @@ static int target_check_cdb_and_preempt(struct list_head *list,
 	return 1;
 }
 
-static bool __target_check_io_state(struct se_cmd *se_cmd)
+static bool __target_check_io_state(struct se_cmd *se_cmd,
+				    struct se_session *tmr_sess, int tas)
 {
 	struct se_session *sess = se_cmd->se_sess;
 
@@ -122,21 +125,33 @@ static bool __target_check_io_state(struct se_cmd *se_cmd)
 	WARN_ON_ONCE(!irqs_disabled());
 	/*
 	 * If command already reached CMD_T_COMPLETE state within
-	 * target_complete_cmd(), this se_cmd has been passed to
-	 * fabric driver and will not be aborted.
+	 * target_complete_cmd() or CMD_T_FABRIC_STOP due to shutdown,
+	 * this se_cmd has been passed to fabric driver and will
+	 * not be aborted.
 	 *
 	 * Otherwise, obtain a local se_cmd->cmd_kref now for TMR
 	 * ABORT_TASK + LUN_RESET for CMD_T_ABORTED processing as
 	 * long as se_cmd->cmd_kref is still active unless zero.
 	 */
 	spin_lock(&se_cmd->t_state_lock);
-	if (se_cmd->transport_state & CMD_T_COMPLETE) {
-		pr_debug("Attempted to abort io tag: %u already complete,"
+	if (se_cmd->transport_state & (CMD_T_COMPLETE | CMD_T_FABRIC_STOP)) {
+		pr_debug("Attempted to abort io tag: %u already complete or"
+			" fabric stop, skipping\n",
+			se_cmd->se_tfo->get_task_tag(se_cmd));
+		spin_unlock(&se_cmd->t_state_lock);
+		return false;
+	}
+	if (sess->sess_tearing_down || se_cmd->cmd_wait_set) {
+		pr_debug("Attempted to abort io tag: %u already shutdown,"
 			" skipping\n", se_cmd->se_tfo->get_task_tag(se_cmd));
 		spin_unlock(&se_cmd->t_state_lock);
 		return false;
 	}
 	se_cmd->transport_state |= CMD_T_ABORTED;
+
+	if ((tmr_sess != se_cmd->se_sess) && tas)
+		se_cmd->transport_state |= CMD_T_TAS;
+
 	spin_unlock(&se_cmd->t_state_lock);
 
 	return kref_get_unless_zero(&se_cmd->cmd_kref);
@@ -169,7 +184,7 @@ void core_tmr_abort_task(
 		printk("ABORT_TASK: Found referenced %s task_tag: %u\n",
 			se_cmd->se_tfo->get_fabric_name(), ref_tag);
 
-		if (!__target_check_io_state(se_cmd)) {
+		if (!__target_check_io_state(se_cmd, se_sess, 0)) {
 			spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
 			target_put_sess_cmd(se_sess, se_cmd);
 			goto out;
@@ -239,7 +254,8 @@ static void core_tmr_drain_tmr_list(
 
 		spin_lock(&sess->sess_cmd_lock);
 		spin_lock(&cmd->t_state_lock);
-		if (!(cmd->transport_state & CMD_T_ACTIVE)) {
+		if (!(cmd->transport_state & CMD_T_ACTIVE) ||
+		     (cmd->transport_state & CMD_T_FABRIC_STOP)) {
 			spin_unlock(&cmd->t_state_lock);
 			spin_unlock(&sess->sess_cmd_lock);
 			continue;
@@ -249,15 +265,22 @@ static void core_tmr_drain_tmr_list(
 			spin_unlock(&sess->sess_cmd_lock);
 			continue;
 		}
+		if (sess->sess_tearing_down || cmd->cmd_wait_set) {
+			spin_unlock(&cmd->t_state_lock);
+			spin_unlock(&sess->sess_cmd_lock);
+			continue;
+		}
 		cmd->transport_state |= CMD_T_ABORTED;
 		spin_unlock(&cmd->t_state_lock);
 
 		rc = kref_get_unless_zero(&cmd->cmd_kref);
-		spin_unlock(&sess->sess_cmd_lock);
 		if (!rc) {
 			printk("LUN_RESET TMR: non-zero kref_get_unless_zero\n");
+			spin_unlock(&sess->sess_cmd_lock);
 			continue;
 		}
+		spin_unlock(&sess->sess_cmd_lock);
+
 		list_move_tail(&tmr_p->tmr_list, &drain_tmr_list);
 	}
 	spin_unlock_irqrestore(&dev->se_tmr_lock, flags);
@@ -334,7 +357,7 @@ static void core_tmr_drain_state_list(
 			continue;
 
 		spin_lock(&sess->sess_cmd_lock);
-		rc = __target_check_io_state(cmd);
+		rc = __target_check_io_state(cmd, tmr_sess, tas);
 		spin_unlock(&sess->sess_cmd_lock);
 		if (!rc)
 			continue;
@@ -373,7 +396,7 @@ static void core_tmr_drain_state_list(
 		cancel_work_sync(&cmd->work);
 		transport_wait_for_tasks(cmd);
 
-		core_tmr_handle_tas_abort(tmr_sess, cmd, tas);
+		core_tmr_handle_tas_abort(cmd, tas);
 		target_put_sess_cmd(cmd->se_sess, cmd);
 	}
 }

drivers/target/target_core_transport.c

@@ -2342,18 +2342,33 @@ static void transport_write_pending_qf(struct se_cmd *cmd)
 	}
 }
 
+static bool
+__transport_wait_for_tasks(struct se_cmd *, bool, bool *, bool *,
+			   unsigned long *flags);
+
+static void target_wait_free_cmd(struct se_cmd *cmd, bool *aborted, bool *tas)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	__transport_wait_for_tasks(cmd, true, aborted, tas, &flags);
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+}
+
 int transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks)
 {
 	int ret = 0;
+	bool aborted = false, tas = false;
 
 	if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD)) {
 		if (wait_for_tasks && (cmd->se_cmd_flags & SCF_SCSI_TMR_CDB))
-			transport_wait_for_tasks(cmd);
+			target_wait_free_cmd(cmd, &aborted, &tas);
 
-		ret = transport_put_cmd(cmd);
+		if (!aborted || tas)
+			ret = transport_put_cmd(cmd);
 	} else {
 		if (wait_for_tasks)
-			transport_wait_for_tasks(cmd);
+			target_wait_free_cmd(cmd, &aborted, &tas);
 		/*
 		 * Handle WRITE failure case where transport_generic_new_cmd()
 		 * has already added se_cmd to state_list, but fabric has
@@ -2365,7 +2380,21 @@ int transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks)
 		if (cmd->se_lun)
 			transport_lun_remove_cmd(cmd);
 
-		ret = transport_put_cmd(cmd);
+		if (!aborted || tas)
+			ret = transport_put_cmd(cmd);
+	}
+	/*
+	 * If the task has been internally aborted due to TMR ABORT_TASK
+	 * or LUN_RESET, target_core_tmr.c is responsible for performing
+	 * the remaining calls to target_put_sess_cmd(), and not the
+	 * callers of this function.
+	 */
+	if (aborted) {
+		pr_debug("Detected CMD_T_ABORTED for ITT: %u\n",
+			cmd->se_tfo->get_task_tag(cmd));
+		wait_for_completion(&cmd->cmd_wait_comp);
+		cmd->se_tfo->release_cmd(cmd);
+		ret = 1;
 	}
 	return ret;
 }
@@ -2422,6 +2451,7 @@ static void target_release_cmd_kref(struct kref *kref)
 {
 	struct se_cmd *se_cmd = container_of(kref, struct se_cmd, cmd_kref);
 	struct se_session *se_sess = se_cmd->se_sess;
+	bool fabric_stop;
 
 	if (list_empty(&se_cmd->se_cmd_list)) {
 		spin_unlock(&se_sess->sess_cmd_lock);
@@ -2429,13 +2459,19 @@ static void target_release_cmd_kref(struct kref *kref)
 		se_cmd->se_tfo->release_cmd(se_cmd);
 		return;
 	}
-	if (se_sess->sess_tearing_down && se_cmd->cmd_wait_set) {
+
+	spin_lock(&se_cmd->t_state_lock);
+	fabric_stop = (se_cmd->transport_state & CMD_T_FABRIC_STOP);
+	spin_unlock(&se_cmd->t_state_lock);
+
+	if (se_cmd->cmd_wait_set || fabric_stop) {
+		list_del_init(&se_cmd->se_cmd_list);
 		spin_unlock(&se_sess->sess_cmd_lock);
 		target_free_cmd_mem(se_cmd);
 		complete(&se_cmd->cmd_wait_comp);
 		return;
 	}
-	list_del(&se_cmd->se_cmd_list);
+	list_del_init(&se_cmd->se_cmd_list);
 	spin_unlock(&se_sess->sess_cmd_lock);
 
 	target_free_cmd_mem(se_cmd);
@@ -2467,6 +2503,7 @@ void target_sess_cmd_list_set_waiting(struct se_session *se_sess)
 {
 	struct se_cmd *se_cmd;
 	unsigned long flags;
+	int rc;
 
 	spin_lock_irqsave(&se_sess->sess_cmd_lock, flags);
 	if (se_sess->sess_tearing_down) {
@@ -2476,8 +2513,15 @@ void target_sess_cmd_list_set_waiting(struct se_session *se_sess)
 	se_sess->sess_tearing_down = 1;
 	list_splice_init(&se_sess->sess_cmd_list, &se_sess->sess_wait_list);
 
-	list_for_each_entry(se_cmd, &se_sess->sess_wait_list, se_cmd_list)
-		se_cmd->cmd_wait_set = 1;
+	list_for_each_entry(se_cmd, &se_sess->sess_wait_list, se_cmd_list) {
+		rc = kref_get_unless_zero(&se_cmd->cmd_kref);
+		if (rc) {
+			se_cmd->cmd_wait_set = 1;
+			spin_lock(&se_cmd->t_state_lock);
+			se_cmd->transport_state |= CMD_T_FABRIC_STOP;
+			spin_unlock(&se_cmd->t_state_lock);
+		}
+	}
 
 	spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
 }
@@ -2490,15 +2534,25 @@ void target_wait_for_sess_cmds(struct se_session *se_sess)
 {
 	struct se_cmd *se_cmd, *tmp_cmd;
 	unsigned long flags;
+	bool tas;
 
 	list_for_each_entry_safe(se_cmd, tmp_cmd,
 				&se_sess->sess_wait_list, se_cmd_list) {
-		list_del(&se_cmd->se_cmd_list);
+		list_del_init(&se_cmd->se_cmd_list);
 
 		pr_debug("Waiting for se_cmd: %p t_state: %d, fabric state:"
 			" %d\n", se_cmd, se_cmd->t_state,
 			se_cmd->se_tfo->get_cmd_state(se_cmd));
 
+		spin_lock_irqsave(&se_cmd->t_state_lock, flags);
+		tas = (se_cmd->transport_state & CMD_T_TAS);
+		spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
+
+		if (!target_put_sess_cmd(se_sess, se_cmd)) {
+			if (tas)
+				target_put_sess_cmd(se_sess, se_cmd);
+		}
+
 		wait_for_completion(&se_cmd->cmd_wait_comp);
 		pr_debug("After cmd_wait_comp: se_cmd: %p t_state: %d"
 			" fabric state: %d\n", se_cmd, se_cmd->t_state,
@@ -2541,34 +2595,38 @@ int transport_clear_lun_ref(struct se_lun *lun)
 	return 0;
 }
 
-/**
- * transport_wait_for_tasks - wait for completion to occur
- * @cmd:	command to wait
- *
- * Called from frontend fabric context to wait for storage engine
- * to pause and/or release frontend generated struct se_cmd.
- */
-bool transport_wait_for_tasks(struct se_cmd *cmd)
+static bool
+__transport_wait_for_tasks(struct se_cmd *cmd, bool fabric_stop,
+			   bool *aborted, bool *tas, unsigned long *flags)
+	__releases(&cmd->t_state_lock)
+	__acquires(&cmd->t_state_lock)
 {
-	unsigned long flags;
 
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	assert_spin_locked(&cmd->t_state_lock);
+	WARN_ON_ONCE(!irqs_disabled());
+
+	if (fabric_stop)
+		cmd->transport_state |= CMD_T_FABRIC_STOP;
+
+	if (cmd->transport_state & CMD_T_ABORTED)
+		*aborted = true;
+
+	if (cmd->transport_state & CMD_T_TAS)
+		*tas = true;
+
 	if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD) &&
-	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB)) {
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB))
 		return false;
-	}
 
 	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);
+	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB))
 		return false;
-	}
 
-	if (!(cmd->transport_state & CMD_T_ACTIVE)) {
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	if (!(cmd->transport_state & CMD_T_ACTIVE))
+		return false;
+
+	if (fabric_stop && *aborted)
 		return false;
-	}
 
 	cmd->transport_state |= CMD_T_STOP;
 
@@ -2577,20 +2635,37 @@ bool transport_wait_for_tasks(struct se_cmd *cmd)
 		cmd, cmd->se_tfo->get_task_tag(cmd),
 		cmd->se_tfo->get_cmd_state(cmd), cmd->t_state);
 
-	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	spin_unlock_irqrestore(&cmd->t_state_lock, *flags);
 
 	wait_for_completion(&cmd->t_transport_stop_comp);
 
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	spin_lock_irqsave(&cmd->t_state_lock, *flags);
 	cmd->transport_state &= ~(CMD_T_ACTIVE | CMD_T_STOP);
 
 	pr_debug("wait_for_tasks: Stopped wait_for_completion("
 		"&cmd->t_transport_stop_comp) for ITT: 0x%08x\n",
 		cmd->se_tfo->get_task_tag(cmd));
 
+	return true;
+}
+
+/**
+ * transport_wait_for_tasks - wait for completion to occur
+ * @cmd:	command to wait
+ *
+ * Called from frontend fabric context to wait for storage engine
+ * to pause and/or release frontend generated struct se_cmd.
+ */
+bool transport_wait_for_tasks(struct se_cmd *cmd)
+{
+	unsigned long flags;
+	bool ret, aborted = false, tas = false;
+
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	ret = __transport_wait_for_tasks(cmd, false, &aborted, &tas, &flags);
 	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
-	return true;
+	return ret;
 }
 EXPORT_SYMBOL(transport_wait_for_tasks);
 

include/target/target_core_base.h

@@ -533,6 +533,8 @@ struct se_cmd {
 #define CMD_T_DEV_ACTIVE	(1 << 7)
 #define CMD_T_REQUEST_STOP	(1 << 8)
 #define CMD_T_BUSY		(1 << 9)
+#define CMD_T_TAS		(1 << 10)
+#define CMD_T_FABRIC_STOP	(1 << 11)
 	spinlock_t		t_state_lock;
 	struct completion	t_transport_stop_comp;