SCSI fixes on 20201120

Fixes for two fairly obscure but annoying when triggered races in
 iSCSI.
 
 Signed-off-by: James E.J. Bottomley <jejb@linux.ibm.com>
 -----BEGIN PGP SIGNATURE-----
 
 iJwEABMIAEQWIQTnYEDbdso9F2cI+arnQslM7pishQUCX7gmiiYcamFtZXMuYm90
 dG9tbGV5QGhhbnNlbnBhcnRuZXJzaGlwLmNvbQAKCRDnQslM7pishcJeAP4v9I6j
 fh/0E3tU+Hb8ZerNoD8p2q7vdxeKkihXWw5/EwEA2SoFW6VNTrJdhUq1+hJ1QWOR
 Awrex3uy1m11ETLgIcM=
 =A1lM
 -----END PGP SIGNATURE-----

Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "Fixes for two fairly obscure but annoying when triggered races in
  iSCSI"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: target: iscsi: Fix cmd abort fabric stop race
  scsi: libiscsi: Fix NOP race condition

drivers/scsi/libiscsi.c

@@ -533,8 +533,8 @@ static void iscsi_complete_task(struct iscsi_task *task, int state)
 	if (conn->task == task)
 		conn->task = NULL;
 
-	if (conn->ping_task == task)
-		conn->ping_task = NULL;
+	if (READ_ONCE(conn->ping_task) == task)
+		WRITE_ONCE(conn->ping_task, NULL);
 
 	/* release get from queueing */
 	__iscsi_put_task(task);
@@ -738,6 +738,9 @@ __iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 						   task->conn->session->age);
 	}
 
+	if (unlikely(READ_ONCE(conn->ping_task) == INVALID_SCSI_TASK))
+		WRITE_ONCE(conn->ping_task, task);
+
 	if (!ihost->workq) {
 		if (iscsi_prep_mgmt_task(conn, task))
 			goto free_task;
@@ -941,8 +944,11 @@ static int iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr)
         struct iscsi_nopout hdr;
 	struct iscsi_task *task;
 
-	if (!rhdr && conn->ping_task)
-		return -EINVAL;
+	if (!rhdr) {
+		if (READ_ONCE(conn->ping_task))
+			return -EINVAL;
+		WRITE_ONCE(conn->ping_task, INVALID_SCSI_TASK);
+	}
 
 	memset(&hdr, 0, sizeof(struct iscsi_nopout));
 	hdr.opcode = ISCSI_OP_NOOP_OUT | ISCSI_OP_IMMEDIATE;
@@ -957,11 +963,12 @@ static int iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr)
 
 	task = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)&hdr, NULL, 0);
 	if (!task) {
+		if (!rhdr)
+			WRITE_ONCE(conn->ping_task, NULL);
 		iscsi_conn_printk(KERN_ERR, conn, "Could not send nopout\n");
 		return -EIO;
 	} else if (!rhdr) {
 		/* only track our nops */
-		conn->ping_task = task;
 		conn->last_ping = jiffies;
 	}
 
@@ -984,7 +991,7 @@ static int iscsi_nop_out_rsp(struct iscsi_task *task,
 	struct iscsi_conn *conn = task->conn;
 	int rc = 0;
 
-	if (conn->ping_task != task) {
+	if (READ_ONCE(conn->ping_task) != task) {
 		/*
 		 * If this is not in response to one of our
 		 * nops then it must be from userspace.
@@ -1923,7 +1930,7 @@ static void iscsi_start_tx(struct iscsi_conn *conn)
  */
 static int iscsi_has_ping_timed_out(struct iscsi_conn *conn)
 {
-	if (conn->ping_task &&
+	if (READ_ONCE(conn->ping_task) &&
 	    time_before_eq(conn->last_recv + (conn->recv_timeout * HZ) +
 			   (conn->ping_timeout * HZ), jiffies))
 		return 1;
@@ -2058,7 +2065,7 @@ enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc)
 	 * Checking the transport already or nop from a cmd timeout still
 	 * running
 	 */
-	if (conn->ping_task) {
+	if (READ_ONCE(conn->ping_task)) {
 		task->have_checked_conn = true;
 		rc = BLK_EH_RESET_TIMER;
 		goto done;

drivers/target/iscsi/iscsi_target.c

@@ -483,8 +483,7 @@ EXPORT_SYMBOL(iscsit_queue_rsp);
 void iscsit_aborted_task(struct iscsi_conn *conn, struct iscsi_cmd *cmd)
 {
 	spin_lock_bh(&conn->cmd_lock);
-	if (!list_empty(&cmd->i_conn_node) &&
-	    !(cmd->se_cmd.transport_state & CMD_T_FABRIC_STOP))
+	if (!list_empty(&cmd->i_conn_node))
 		list_del_init(&cmd->i_conn_node);
 	spin_unlock_bh(&conn->cmd_lock);
 
@@ -4083,12 +4082,22 @@ static void iscsit_release_commands_from_conn(struct iscsi_conn *conn)
 	spin_lock_bh(&conn->cmd_lock);
 	list_splice_init(&conn->conn_cmd_list, &tmp_list);
 
-	list_for_each_entry(cmd, &tmp_list, i_conn_node) {
+	list_for_each_entry_safe(cmd, cmd_tmp, &tmp_list, i_conn_node) {
 		struct se_cmd *se_cmd = &cmd->se_cmd;
 
 		if (se_cmd->se_tfo != NULL) {
 			spin_lock_irq(&se_cmd->t_state_lock);
-			se_cmd->transport_state |= CMD_T_FABRIC_STOP;
+			if (se_cmd->transport_state & CMD_T_ABORTED) {
+				/*
+				 * LIO's abort path owns the cleanup for this,
+				 * so put it back on the list and let
+				 * aborted_task handle it.
+				 */
+				list_move_tail(&cmd->i_conn_node,
+					       &conn->conn_cmd_list);
+			} else {
+				se_cmd->transport_state |= CMD_T_FABRIC_STOP;
+			}
 			spin_unlock_irq(&se_cmd->t_state_lock);
 		}
 	}

include/scsi/libiscsi.h

@@ -132,6 +132,9 @@ struct iscsi_task {
 	void			*dd_data;	/* driver/transport data */
 };
 
+/* invalid scsi_task pointer */
+#define	INVALID_SCSI_TASK	(struct iscsi_task *)-1l
+
 static inline int iscsi_task_has_unsol_data(struct iscsi_task *task)
 {
 	return task->unsol_r2t.data_length > task->unsol_r2t.sent;