Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md

Pull MD updates from Shaohua Li:
 "Some small fixes for MD:

   - fix raid5-cache potential problems if raid5 cache isn't fully
     recovered

   - fix a wait-within-wait warning in raid1/10

   - make raid5-PPL support disks with writeback cache enabled"

* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
  raid5-ppl: PPL support for disks with write-back cache enabled
  md/r5cache: print more info of log recovery
  md/raid1,raid10: silence warning about wait-within-wait
  md: introduce new personality funciton start()

Author: torvalds

Documentation/md/raid5-ppl.txt

@@ -39,6 +39,7 @@ case the behavior is the same as in plain raid5.
 PPL is available for md version-1 metadata and external (specifically IMSM)
 metadata arrays. It can be enabled using mdadm option --consistency-policy=ppl.
 
-Currently, volatile write-back cache should be disabled on all member drives
-when using PPL. Otherwise it cannot guarantee consistency in case of power
-failure.
+There is a limitation of maximum 64 disks in the array for PPL. It allows to
+keep data structures and implementation simple. RAID5 arrays with so many disks
+are not likely due to high risk of multiple disks failure. Such restriction
+should not be a real life limitation.

drivers/md/dm-raid.c

@@ -3151,6 +3151,14 @@ static int raid_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		goto bad;
 	}
 
+	r = md_start(&rs->md);
+
+	if (r) {
+		ti->error = "Failed to start raid array";
+		mddev_unlock(&rs->md);
+		goto bad_md_start;
+	}
+
 	rs->callbacks.congested_fn = raid_is_congested;
 	dm_table_add_target_callbacks(ti->table, &rs->callbacks);
 
@@ -3198,6 +3206,7 @@ static int raid_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	mddev_unlock(&rs->md);
 	return 0;
 
+bad_md_start:
 bad_journal_mode_set:
 bad_stripe_cache:
 bad_check_reshape:

drivers/md/md.c

@@ -711,7 +711,7 @@ static struct md_rdev *find_rdev(struct mddev *mddev, dev_t dev)
 	return NULL;
 }
 
-static struct md_rdev *find_rdev_rcu(struct mddev *mddev, dev_t dev)
+struct md_rdev *md_find_rdev_rcu(struct mddev *mddev, dev_t dev)
 {
 	struct md_rdev *rdev;
 
@@ -721,6 +721,7 @@ static struct md_rdev *find_rdev_rcu(struct mddev *mddev, dev_t dev)
 
 	return NULL;
 }
+EXPORT_SYMBOL_GPL(md_find_rdev_rcu);
 
 static struct md_personality *find_pers(int level, char *clevel)
 {
@@ -5560,11 +5561,6 @@ int md_run(struct mddev *mddev)
 	if (start_readonly && mddev->ro == 0)
 		mddev->ro = 2; /* read-only, but switch on first write */
 
-	/*
-	 * NOTE: some pers->run(), for example r5l_recovery_log(), wakes
-	 * up mddev->thread. It is important to initialize critical
-	 * resources for mddev->thread BEFORE calling pers->run().
-	 */
 	err = pers->run(mddev);
 	if (err)
 		pr_warn("md: pers->run() failed ...\n");
@@ -5678,6 +5674,9 @@ static int do_md_run(struct mddev *mddev)
 	if (mddev_is_clustered(mddev))
 		md_allow_write(mddev);
 
+	/* run start up tasks that require md_thread */
+	md_start(mddev);
+
 	md_wakeup_thread(mddev->thread);
 	md_wakeup_thread(mddev->sync_thread); /* possibly kick off a reshape */
 
@@ -5689,6 +5688,21 @@ static int do_md_run(struct mddev *mddev)
 	return err;
 }
 
+int md_start(struct mddev *mddev)
+{
+	int ret = 0;
+
+	if (mddev->pers->start) {
+		set_bit(MD_RECOVERY_WAIT, &mddev->recovery);
+		md_wakeup_thread(mddev->thread);
+		ret = mddev->pers->start(mddev);
+		clear_bit(MD_RECOVERY_WAIT, &mddev->recovery);
+		md_wakeup_thread(mddev->sync_thread);
+	}
+	return ret;
+}
+EXPORT_SYMBOL_GPL(md_start);
+
 static int restart_array(struct mddev *mddev)
 {
 	struct gendisk *disk = mddev->gendisk;
@@ -6997,7 +7011,7 @@ static int set_disk_faulty(struct mddev *mddev, dev_t dev)
 		return -ENODEV;
 
 	rcu_read_lock();
-	rdev = find_rdev_rcu(mddev, dev);
+	rdev = md_find_rdev_rcu(mddev, dev);
 	if (!rdev)
 		err =  -ENODEV;
 	else {
@@ -8169,7 +8183,8 @@ void md_do_sync(struct md_thread *thread)
 	int ret;
 
 	/* just incase thread restarts... */
-	if (test_bit(MD_RECOVERY_DONE, &mddev->recovery))
+	if (test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
+	    test_bit(MD_RECOVERY_WAIT, &mddev->recovery))
 		return;
 	if (mddev->ro) {/* never try to sync a read-only array */
 		set_bit(MD_RECOVERY_INTR, &mddev->recovery);

drivers/md/md.h

@@ -485,6 +485,7 @@ enum recovery_flags {
 	MD_RECOVERY_RESHAPE,	/* A reshape is happening */
 	MD_RECOVERY_FROZEN,	/* User request to abort, and not restart, any action */
 	MD_RECOVERY_ERROR,	/* sync-action interrupted because io-error */
+	MD_RECOVERY_WAIT,	/* waiting for pers->start() to finish */
 };
 
 static inline int __must_check mddev_lock(struct mddev *mddev)
@@ -523,7 +524,13 @@ struct md_personality
 	struct list_head list;
 	struct module *owner;
 	bool (*make_request)(struct mddev *mddev, struct bio *bio);
+	/*
+	 * start up works that do NOT require md_thread. tasks that
+	 * requires md_thread should go into start()
+	 */
 	int (*run)(struct mddev *mddev);
+	/* start up works that require md threads */
+	int (*start)(struct mddev *mddev);
 	void (*free)(struct mddev *mddev, void *priv);
 	void (*status)(struct seq_file *seq, struct mddev *mddev);
 	/* error_handler must set ->faulty and clear ->in_sync
@@ -687,6 +694,7 @@ extern int strict_strtoul_scaled(const char *cp, unsigned long *res, int scale);
 
 extern void mddev_init(struct mddev *mddev);
 extern int md_run(struct mddev *mddev);
+extern int md_start(struct mddev *mddev);
 extern void md_stop(struct mddev *mddev);
 extern void md_stop_writes(struct mddev *mddev);
 extern int md_rdev_init(struct md_rdev *rdev);
@@ -702,6 +710,7 @@ extern void md_reload_sb(struct mddev *mddev, int raid_disk);
 extern void md_update_sb(struct mddev *mddev, int force);
 extern void md_kick_rdev_from_array(struct md_rdev * rdev);
 struct md_rdev *md_find_rdev_nr_rcu(struct mddev *mddev, int nr);
+struct md_rdev *md_find_rdev_rcu(struct mddev *mddev, dev_t dev);
 
 static inline void rdev_dec_pending(struct md_rdev *rdev, struct mddev *mddev)
 {

drivers/md/raid1.c

@@ -815,6 +815,17 @@ static void flush_pending_writes(struct r1conf *conf)
 		bio = bio_list_get(&conf->pending_bio_list);
 		conf->pending_count = 0;
 		spin_unlock_irq(&conf->device_lock);
+
+		/*
+		 * As this is called in a wait_event() loop (see freeze_array),
+		 * current->state might be TASK_UNINTERRUPTIBLE which will
+		 * cause a warning when we prepare to wait again.  As it is
+		 * rare that this path is taken, it is perfectly safe to force
+		 * us to go around the wait_event() loop again, so the warning
+		 * is a false-positive.  Silence the warning by resetting
+		 * thread state
+		 */
+		__set_current_state(TASK_RUNNING);
 		blk_start_plug(&plug);
 		flush_bio_list(conf, bio);
 		blk_finish_plug(&plug);

drivers/md/raid10.c

@@ -900,6 +900,18 @@ static void flush_pending_writes(struct r10conf *conf)
 		bio = bio_list_get(&conf->pending_bio_list);
 		conf->pending_count = 0;
 		spin_unlock_irq(&conf->device_lock);
+
+		/*
+		 * As this is called in a wait_event() loop (see freeze_array),
+		 * current->state might be TASK_UNINTERRUPTIBLE which will
+		 * cause a warning when we prepare to wait again.  As it is
+		 * rare that this path is taken, it is perfectly safe to force
+		 * us to go around the wait_event() loop again, so the warning
+		 * is a false-positive. Silence the warning by resetting
+		 * thread state
+		 */
+		__set_current_state(TASK_RUNNING);
+
 		blk_start_plug(&plug);
 		/* flush any pending bitmap writes to disk
 		 * before proceeding w/ I/O */

drivers/md/raid5-cache.c

@@ -1111,9 +1111,6 @@ void r5l_write_stripe_run(struct r5l_log *log)
 
 int r5l_handle_flush_request(struct r5l_log *log, struct bio *bio)
 {
-	if (!log)
-		return -ENODEV;
-
 	if (log->r5c_journal_mode == R5C_JOURNAL_MODE_WRITE_THROUGH) {
 		/*
 		 * in write through (journal only)
@@ -1592,8 +1589,6 @@ void r5l_wake_reclaim(struct r5l_log *log, sector_t space)
 void r5l_quiesce(struct r5l_log *log, int quiesce)
 {
 	struct mddev *mddev;
-	if (!log)
-		return;
 
 	if (quiesce) {
 		/* make sure r5l_write_super_and_discard_space exits */
@@ -2448,7 +2443,6 @@ static void r5c_recovery_flush_data_only_stripes(struct r5l_log *log,
 		raid5_release_stripe(sh);
 	}
 
-	md_wakeup_thread(conf->mddev->thread);
 	/* reuse conf->wait_for_quiescent in recovery */
 	wait_event(conf->wait_for_quiescent,
 		   atomic_read(&conf->active_stripes) == 0);
@@ -2491,10 +2485,10 @@ static int r5l_recovery_log(struct r5l_log *log)
 	ctx->seq += 10000;
 
 	if ((ctx->data_only_stripes == 0) && (ctx->data_parity_stripes == 0))
-		pr_debug("md/raid:%s: starting from clean shutdown\n",
+		pr_info("md/raid:%s: starting from clean shutdown\n",
 			 mdname(mddev));
 	else
-		pr_debug("md/raid:%s: recovering %d data-only stripes and %d data-parity stripes\n",
+		pr_info("md/raid:%s: recovering %d data-only stripes and %d data-parity stripes\n",
 			 mdname(mddev), ctx->data_only_stripes,
 			 ctx->data_parity_stripes);
 
@@ -3036,6 +3030,23 @@ static int r5l_load_log(struct r5l_log *log)
 	return ret;
 }
 
+int r5l_start(struct r5l_log *log)
+{
+	int ret;
+
+	if (!log)
+		return 0;
+
+	ret = r5l_load_log(log);
+	if (ret) {
+		struct mddev *mddev = log->rdev->mddev;
+		struct r5conf *conf = mddev->private;
+
+		r5l_exit_log(conf);
+	}
+	return ret;
+}
+
 void r5c_update_on_rdev_error(struct mddev *mddev, struct md_rdev *rdev)
 {
 	struct r5conf *conf = mddev->private;
@@ -3138,13 +3149,9 @@ int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev)
 
 	rcu_assign_pointer(conf->log, log);
 
-	if (r5l_load_log(log))
-		goto error;
-
 	set_bit(MD_HAS_JOURNAL, &conf->mddev->flags);
 	return 0;
 
-error:
 	rcu_assign_pointer(conf->log, NULL);
 	md_unregister_thread(&log->reclaim_thread);
 reclaim_thread:

drivers/md/raid5-log.h

@@ -32,6 +32,7 @@ extern struct md_sysfs_entry r5c_journal_mode;
 extern void r5c_update_on_rdev_error(struct mddev *mddev,
 				     struct md_rdev *rdev);
 extern bool r5c_big_stripe_cached(struct r5conf *conf, sector_t sect);
+extern int r5l_start(struct r5l_log *log);
 
 extern struct dma_async_tx_descriptor *
 ops_run_partial_parity(struct stripe_head *sh, struct raid5_percpu *percpu,
@@ -42,6 +43,7 @@ extern int ppl_write_stripe(struct r5conf *conf, struct stripe_head *sh);
 extern void ppl_write_stripe_run(struct r5conf *conf);
 extern void ppl_stripe_write_finished(struct stripe_head *sh);
 extern int ppl_modify_log(struct r5conf *conf, struct md_rdev *rdev, bool add);
+extern void ppl_quiesce(struct r5conf *conf, int quiesce);
 
 static inline bool raid5_has_ppl(struct r5conf *conf)
 {
@@ -87,6 +89,34 @@ static inline void log_write_stripe_run(struct r5conf *conf)
 		ppl_write_stripe_run(conf);
 }
 
+static inline void log_flush_stripe_to_raid(struct r5conf *conf)
+{
+	if (conf->log)
+		r5l_flush_stripe_to_raid(conf->log);
+	else if (raid5_has_ppl(conf))
+		ppl_write_stripe_run(conf);
+}
+
+static inline int log_handle_flush_request(struct r5conf *conf, struct bio *bio)
+{
+	int ret = -ENODEV;
+
+	if (conf->log)
+		ret = r5l_handle_flush_request(conf->log, bio);
+	else if (raid5_has_ppl(conf))
+		ret = 0;
+
+	return ret;
+}
+
+static inline void log_quiesce(struct r5conf *conf, int quiesce)
+{
+	if (conf->log)
+		r5l_quiesce(conf->log, quiesce);
+	else if (raid5_has_ppl(conf))
+		ppl_quiesce(conf, quiesce);
+}
+
 static inline void log_exit(struct r5conf *conf)
 {
 	if (conf->log)