dm mpath: Add timeout mechanism for queue_if_no_path

Add a configurable timeout mechanism to disable queue_if_no_path without
assistance from userspace multipathd.  This reimplements multipathd's
no_path_retry mechanism in kernel space.  This is motivated by the
desire to prevent processes from hanging indefinitely waiting for IO
in cases where multipathd might be unable to respond (after a failure
or for whatever reason).

Despite replicating userspace multipathd's policy configuration in
kernel space, it is important to prevent IOs from hanging forever,
waiting for userspace that may be incapable of behaving correctly.

Use of the provided "queue_if_no_path_timeout_secs" dm-multipath
module parameter is optional.  This timeout mechanism is disabled by
default (by being set to 0).
Signed-off-by: Anatol Pomazau <anatol@google.com>
Co-developed-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>

drivers/md/dm-mpath.c

@@ -20,6 +20,7 @@
 #include <linux/pagemap.h>
 #include <linux/slab.h>
 #include <linux/time.h>
+#include <linux/timer.h>
 #include <linux/workqueue.h>
 #include <linux/delay.h>
 #include <scsi/scsi_dh.h>
@@ -29,6 +30,9 @@
 #define DM_MSG_PREFIX "multipath"
 #define DM_PG_INIT_DELAY_MSECS 2000
 #define DM_PG_INIT_DELAY_DEFAULT ((unsigned) -1)
+#define QUEUE_IF_NO_PATH_TIMEOUT_DEFAULT 0
+
+static unsigned long queue_if_no_path_timeout_secs = QUEUE_IF_NO_PATH_TIMEOUT_DEFAULT;
 
 /* Path properties */
 struct pgpath {
@@ -91,6 +95,8 @@ struct multipath {
 
 	struct work_struct process_queued_bios;
 	struct bio_list queued_bios;
+
+	struct timer_list nopath_timer;	/* Timeout for queue_if_no_path */
 };
 
 /*
@@ -108,6 +114,7 @@ static void trigger_event(struct work_struct *work);
 static void activate_or_offline_path(struct pgpath *pgpath);
 static void activate_path_work(struct work_struct *work);
 static void process_queued_bios(struct work_struct *work);
+static void queue_if_no_path_timeout_work(struct timer_list *t);
 
 /*-----------------------------------------------
  * Multipath state flags.
@@ -195,6 +202,8 @@ static struct multipath *alloc_multipath(struct dm_target *ti)
 
 		m->ti = ti;
 		ti->private = m;
+
+		timer_setup(&m->nopath_timer, queue_if_no_path_timeout_work, 0);
 	}
 
 	return m;
@@ -717,6 +726,43 @@ static int queue_if_no_path(struct multipath *m, bool queue_if_no_path,
 	return 0;
 }
 
+/*
+ * If the queue_if_no_path timeout fires, turn off queue_if_no_path and
+ * process any queued I/O.
+ */
+static void queue_if_no_path_timeout_work(struct timer_list *t)
+{
+	struct multipath *m = from_timer(m, t, nopath_timer);
+	struct mapped_device *md = dm_table_get_md(m->ti->table);
+
+	DMWARN("queue_if_no_path timeout on %s, failing queued IO", dm_device_name(md));
+	queue_if_no_path(m, false, false);
+}
+
+/*
+ * Enable the queue_if_no_path timeout if necessary.
+ * Called with m->lock held.
+ */
+static void enable_nopath_timeout(struct multipath *m)
+{
+	unsigned long queue_if_no_path_timeout =
+		READ_ONCE(queue_if_no_path_timeout_secs) * HZ;
+
+	lockdep_assert_held(&m->lock);
+
+	if (queue_if_no_path_timeout > 0 &&
+	    atomic_read(&m->nr_valid_paths) == 0 &&
+	    test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags)) {
+		mod_timer(&m->nopath_timer,
+			  jiffies + queue_if_no_path_timeout);
+	}
+}
+
+static void disable_nopath_timeout(struct multipath *m)
+{
+	del_timer_sync(&m->nopath_timer);
+}
+
 /*
  * An event is triggered whenever a path is taken out of use.
  * Includes path failure and PG bypass.
@@ -1090,6 +1136,7 @@ static int multipath_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	struct dm_arg_set as;
 	unsigned pg_count = 0;
 	unsigned next_pg_num;
+	unsigned long flags;
 
 	as.argc = argc;
 	as.argv = argv;
@@ -1154,6 +1201,10 @@ static int multipath_ctr(struct dm_target *ti, unsigned argc, char **argv)
 		goto bad;
 	}
 
+	spin_lock_irqsave(&m->lock, flags);
+	enable_nopath_timeout(m);
+	spin_unlock_irqrestore(&m->lock, flags);
+
 	ti->num_flush_bios = 1;
 	ti->num_discard_bios = 1;
 	ti->num_write_same_bios = 1;
@@ -1208,6 +1259,7 @@ static void multipath_dtr(struct dm_target *ti)
 {
 	struct multipath *m = ti->private;
 
+	disable_nopath_timeout(m);
 	flush_multipath_work(m);
 	free_multipath(m);
 }
@@ -1241,6 +1293,8 @@ static int fail_path(struct pgpath *pgpath)
 
 	schedule_work(&m->trigger_event);
 
+	enable_nopath_timeout(m);
+
 out:
 	spin_unlock_irqrestore(&m->lock, flags);
 
@@ -1291,6 +1345,9 @@ static int reinstate_path(struct pgpath *pgpath)
 		process_queued_io_list(m);
 	}
 
+	if (pgpath->is_active)
+		disable_nopath_timeout(m);
+
 	return r;
 }
 
@@ -1789,6 +1846,7 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv,
 	struct dm_dev *dev;
 	struct multipath *m = ti->private;
 	action_fn action;
+	unsigned long flags;
 
 	mutex_lock(&m->work_mutex);
 
@@ -1800,9 +1858,13 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv,
 	if (argc == 1) {
 		if (!strcasecmp(argv[0], "queue_if_no_path")) {
 			r = queue_if_no_path(m, true, false);
+			spin_lock_irqsave(&m->lock, flags);
+			enable_nopath_timeout(m);
+			spin_unlock_irqrestore(&m->lock, flags);
 			goto out;
 		} else if (!strcasecmp(argv[0], "fail_if_no_path")) {
 			r = queue_if_no_path(m, false, false);
+			disable_nopath_timeout(m);
 			goto out;
 		}
 	}
@@ -2065,6 +2127,10 @@ static void __exit dm_multipath_exit(void)
 module_init(dm_multipath_init);
 module_exit(dm_multipath_exit);
 
+module_param_named(queue_if_no_path_timeout_secs,
+		   queue_if_no_path_timeout_secs, ulong, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(queue_if_no_path_timeout_secs, "No available paths queue IO timeout in seconds");
+
 MODULE_DESCRIPTION(DM_NAME " multipath target");
 MODULE_AUTHOR("Sistina Software <dm-devel@redhat.com>");
 MODULE_LICENSE("GPL");