Merge tag 'md-3.5-fixes' of git://neil.brown.name/md

Pull md fixes from NeilBrown:
 "md: collection of bug fixes for 3.5

  You go away for 2 weeks vacation and what do you get when you come
  back? Piles of bugs :-)

  Some found by inspection, some by testing, some during use in the
  field, and some while developing for the next window..."

* tag 'md-3.5-fixes' of git://neil.brown.name/md:
  md: fix up plugging (again).
  md: support re-add of recovering devices.
  md/raid1: fix bug in read_balance introduced by hot-replace
  raid5: delayed stripe fix
  md/raid456: When read error cannot be recovered, record bad block
  md: make 'name' arg to md_register_thread non-optional.
  md/raid10: fix failure when trying to repair a read error.
  md/raid5: fix refcount problem when blocked_rdev is set.
  md:Add blk_plug in sync_thread.
  md/raid5: In ops_run_io, inc nr_pending before calling md_wait_for_blocked_rdev
  md/raid5: Do not add data_offset before call to is_badblock
  md/raid5: prefer replacing failed devices over want-replacement devices.
  md/raid10: Don't try to recovery unmatched (and unused) chunks.

drivers/md/md.c

@@ -5784,8 +5784,7 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info)
 			super_types[mddev->major_version].
 				validate_super(mddev, rdev);
 		if ((info->state & (1<<MD_DISK_SYNC)) &&
-		    (!test_bit(In_sync, &rdev->flags) ||
-		     rdev->raid_disk != info->raid_disk)) {
+		     rdev->raid_disk != info->raid_disk) {
 			/* This was a hot-add request, but events doesn't
 			 * match, so reject it.
 			 */
@@ -6751,7 +6750,7 @@ struct md_thread *md_register_thread(void (*run) (struct mddev *), struct mddev
 	thread->tsk = kthread_run(md_thread, thread,
 				  "%s_%s",
 				  mdname(thread->mddev),
-				  name ?: mddev->pers->name);
+				  name);
 	if (IS_ERR(thread->tsk)) {
 		kfree(thread);
 		return NULL;
@@ -7298,6 +7297,7 @@ void md_do_sync(struct mddev *mddev)
 	int skipped = 0;
 	struct md_rdev *rdev;
 	char *desc;
+	struct blk_plug plug;
 
 	/* just incase thread restarts... */
 	if (test_bit(MD_RECOVERY_DONE, &mddev->recovery))
@@ -7447,6 +7447,7 @@ void md_do_sync(struct mddev *mddev)
 	}
 	mddev->curr_resync_completed = j;
 
+	blk_start_plug(&plug);
 	while (j < max_sectors) {
 		sector_t sectors;
 
@@ -7552,6 +7553,7 @@ void md_do_sync(struct mddev *mddev)
 	 * this also signals 'finished resyncing' to md_stop
 	 */
  out:
+	blk_finish_plug(&plug);
 	wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active));
 
 	/* tell personality that we are finished */

drivers/md/multipath.c

@@ -474,7 +474,8 @@ static int multipath_run (struct mddev *mddev)
 	}
 
 	{
-		mddev->thread = md_register_thread(multipathd, mddev, NULL);
+		mddev->thread = md_register_thread(multipathd, mddev,
+						   "multipath");
 		if (!mddev->thread) {
 			printk(KERN_ERR "multipath: couldn't allocate thread"
 				" for %s\n", mdname(mddev));

drivers/md/raid1.c

@@ -517,8 +517,8 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
 		int bad_sectors;
 
 		int disk = start_disk + i;
-		if (disk >= conf->raid_disks)
-			disk -= conf->raid_disks;
+		if (disk >= conf->raid_disks * 2)
+			disk -= conf->raid_disks * 2;
 
 		rdev = rcu_dereference(conf->mirrors[disk].rdev);
 		if (r1_bio->bios[disk] == IO_BLOCKED
@@ -883,7 +883,6 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	const unsigned long do_sync = (bio->bi_rw & REQ_SYNC);
 	const unsigned long do_flush_fua = (bio->bi_rw & (REQ_FLUSH | REQ_FUA));
 	struct md_rdev *blocked_rdev;
-	int plugged;
 	int first_clone;
 	int sectors_handled;
 	int max_sectors;
@@ -1034,7 +1033,6 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	 * the bad blocks.  Each set of writes gets it's own r1bio
 	 * with a set of bios attached.
 	 */
-	plugged = mddev_check_plugged(mddev);
 
 	disks = conf->raid_disks * 2;
  retry_write:
@@ -1191,6 +1189,8 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 		bio_list_add(&conf->pending_bio_list, mbio);
 		conf->pending_count++;
 		spin_unlock_irqrestore(&conf->device_lock, flags);
+		if (!mddev_check_plugged(mddev))
+			md_wakeup_thread(mddev->thread);
 	}
 	/* Mustn't call r1_bio_write_done before this next test,
 	 * as it could result in the bio being freed.
@@ -1213,9 +1213,6 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 
 	/* In case raid1d snuck in to freeze_array */
 	wake_up(&conf->wait_barrier);
-
-	if (do_sync || !bitmap || !plugged)
-		md_wakeup_thread(mddev->thread);
 }
 
 static void status(struct seq_file *seq, struct mddev *mddev)
@@ -2621,7 +2618,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 		goto abort;
 	}
 	err = -ENOMEM;
-	conf->thread = md_register_thread(raid1d, mddev, NULL);
+	conf->thread = md_register_thread(raid1d, mddev, "raid1");
 	if (!conf->thread) {
 		printk(KERN_ERR
 		       "md/raid1:%s: couldn't allocate thread\n",

drivers/md/raid10.c

@@ -1039,7 +1039,6 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	const unsigned long do_fua = (bio->bi_rw & REQ_FUA);
 	unsigned long flags;
 	struct md_rdev *blocked_rdev;
-	int plugged;
 	int sectors_handled;
 	int max_sectors;
 	int sectors;
@@ -1239,7 +1238,6 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	 * of r10_bios is recored in bio->bi_phys_segments just as with
 	 * the read case.
 	 */
-	plugged = mddev_check_plugged(mddev);
 
 	r10_bio->read_slot = -1; /* make sure repl_bio gets freed */
 	raid10_find_phys(conf, r10_bio);
@@ -1396,6 +1394,8 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 		bio_list_add(&conf->pending_bio_list, mbio);
 		conf->pending_count++;
 		spin_unlock_irqrestore(&conf->device_lock, flags);
+		if (!mddev_check_plugged(mddev, 0, 0))
+			md_wakeup_thread(mddev->thread);
 
 		if (!r10_bio->devs[i].repl_bio)
 			continue;
@@ -1423,6 +1423,8 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 		bio_list_add(&conf->pending_bio_list, mbio);
 		conf->pending_count++;
 		spin_unlock_irqrestore(&conf->device_lock, flags);
+		if (!mddev_check_plugged(mddev))
+			md_wakeup_thread(mddev->thread);
 	}
 
 	/* Don't remove the bias on 'remaining' (one_write_done) until
@@ -1448,9 +1450,6 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 
 	/* In case raid10d snuck in to freeze_array */
 	wake_up(&conf->wait_barrier);
-
-	if (do_sync || !mddev->bitmap || !plugged)
-		md_wakeup_thread(mddev->thread);
 }
 
 static void status(struct seq_file *seq, struct mddev *mddev)
@@ -2310,7 +2309,7 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10
 			if (r10_sync_page_io(rdev,
 					     r10_bio->devs[sl].addr +
 					     sect,
-					     s<<9, conf->tmppage, WRITE)
+					     s, conf->tmppage, WRITE)
 			    == 0) {
 				/* Well, this device is dead */
 				printk(KERN_NOTICE
@@ -2349,7 +2348,7 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10
 			switch (r10_sync_page_io(rdev,
 					     r10_bio->devs[sl].addr +
 					     sect,
-					     s<<9, conf->tmppage,
+					     s, conf->tmppage,
 						 READ)) {
 			case 0:
 				/* Well, this device is dead */
@@ -2512,7 +2511,7 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio)
 	slot = r10_bio->read_slot;
 	printk_ratelimited(
 		KERN_ERR
-		"md/raid10:%s: %s: redirecting"
+		"md/raid10:%s: %s: redirecting "
 		"sector %llu to another mirror\n",
 		mdname(mddev),
 		bdevname(rdev->bdev, b),
@@ -2661,7 +2660,8 @@ static void raid10d(struct mddev *mddev)
 	blk_start_plug(&plug);
 	for (;;) {
 
-		flush_pending_writes(conf);
+		if (atomic_read(&mddev->plug_cnt) == 0)
+			flush_pending_writes(conf);
 
 		spin_lock_irqsave(&conf->device_lock, flags);
 		if (list_empty(head)) {
@@ -2890,6 +2890,12 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr,
 			/* want to reconstruct this device */
 			rb2 = r10_bio;
 			sect = raid10_find_virt(conf, sector_nr, i);
+			if (sect >= mddev->resync_max_sectors) {
+				/* last stripe is not complete - don't
+				 * try to recover this sector.
+				 */
+				continue;
+			}
 			/* Unless we are doing a full sync, or a replacement
 			 * we only need to recover the block if it is set in
 			 * the bitmap
@@ -3421,7 +3427,7 @@ static struct r10conf *setup_conf(struct mddev *mddev)
 	spin_lock_init(&conf->resync_lock);
 	init_waitqueue_head(&conf->wait_barrier);
 
-	conf->thread = md_register_thread(raid10d, mddev, NULL);
+	conf->thread = md_register_thread(raid10d, mddev, "raid10");
 	if (!conf->thread)
 		goto out;