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.

5 files changed, 75 insertions, 42 deletions

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 1c2f9048e1ae..a4c219e3c859 100644
--- a/drivers/md/md.c
+++ b/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 */
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index 9339e67fcc79..61a1833ebaf3 100644
--- a/drivers/md/multipath.c
+++ b/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));
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index a9c7981ddd24..8c2754f835ef 100644
--- a/drivers/md/raid1.c
+++ b/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 @@ read_again:
 	 * 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 @@ read_again:
 		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 @@ read_again:
 
 	/* 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",
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 99ae6068e456..acf5a828c7e1 100644
--- a/drivers/md/raid10.c
+++ b/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 @@ read_again:
 	 * 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 @@ retry_write:
 		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 @@ retry_write:
 		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 @@ retry_write:
 
 	/* 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 @@ read_more:
 	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;
 
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index d26767246d26..04348d76bb30 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -196,12 +196,14 @@ static void __release_stripe(struct r5conf *conf, struct stripe_head *sh)
 		BUG_ON(!list_empty(&sh->lru));
 		BUG_ON(atomic_read(&conf->active_stripes)==0);
 		if (test_bit(STRIPE_HANDLE, &sh->state)) {
-			if (test_bit(STRIPE_DELAYED, &sh->state))
+			if (test_bit(STRIPE_DELAYED, &sh->state) &&
+			    !test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
 				list_add_tail(&sh->lru, &conf->delayed_list);
 			else if (test_bit(STRIPE_BIT_DELAY, &sh->state) &&
 				   sh->bm_seq - conf->seq_write > 0)
 				list_add_tail(&sh->lru, &conf->bitmap_list);
 			else {
+				clear_bit(STRIPE_DELAYED, &sh->state);
 				clear_bit(STRIPE_BIT_DELAY, &sh->state);
 				list_add_tail(&sh->lru, &conf->handle_list);
 			}
@@ -606,6 +608,12 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 					 * a chance*/
 					md_check_recovery(conf->mddev);
 				}
+				/*
+				 * Because md_wait_for_blocked_rdev
+				 * will dec nr_pending, we must
+				 * increment it first.
+				 */
+				atomic_inc(&rdev->nr_pending);
 				md_wait_for_blocked_rdev(rdev, conf->mddev);
 			} else {
 				/* Acknowledged bad block - skip the write */
@@ -1737,6 +1745,7 @@ static void raid5_end_read_request(struct bio * bi, int error)
 	} else {
 		const char *bdn = bdevname(rdev->bdev, b);
 		int retry = 0;
+		int set_bad = 0;
 
 		clear_bit(R5_UPTODATE, &sh->dev[i].flags);
 		atomic_inc(&rdev->read_errors);
@@ -1748,7 +1757,8 @@ static void raid5_end_read_request(struct bio * bi, int error)
 				mdname(conf->mddev),
 				(unsigned long long)s,
 				bdn);
-		else if (conf->mddev->degraded >= conf->max_degraded)
+		else if (conf->mddev->degraded >= conf->max_degraded) {
+			set_bad = 1;
 			printk_ratelimited(
 				KERN_WARNING
 				"md/raid:%s: read error not correctable "
@@ -1756,8 +1766,9 @@ static void raid5_end_read_request(struct bio * bi, int error)
 				mdname(conf->mddev),
 				(unsigned long long)s,
 				bdn);
-		else if (test_bit(R5_ReWrite, &sh->dev[i].flags))
+		} else if (test_bit(R5_ReWrite, &sh->dev[i].flags)) {
 			/* Oh, no!!! */
+			set_bad = 1;
 			printk_ratelimited(
 				KERN_WARNING
 				"md/raid:%s: read error NOT corrected!! "
@@ -1765,7 +1776,7 @@ static void raid5_end_read_request(struct bio * bi, int error)
 				mdname(conf->mddev),
 				(unsigned long long)s,
 				bdn);
-		else if (atomic_read(&rdev->read_errors)
+		} else if (atomic_read(&rdev->read_errors)
 			 > conf->max_nr_stripes)
 			printk(KERN_WARNING
 			       "md/raid:%s: Too many read errors, failing device %s.\n",
@@ -1777,7 +1788,11 @@ static void raid5_end_read_request(struct bio * bi, int error)
 		else {
 			clear_bit(R5_ReadError, &sh->dev[i].flags);
 			clear_bit(R5_ReWrite, &sh->dev[i].flags);
-			md_error(conf->mddev, rdev);
+			if (!(set_bad
+			      && test_bit(In_sync, &rdev->flags)
+			      && rdev_set_badblocks(
+				      rdev, sh->sector, STRIPE_SECTORS, 0)))
+				md_error(conf->mddev, rdev);
 		}
 	}
 	rdev_dec_pending(rdev, conf->mddev);
@@ -3582,8 +3597,18 @@ static void handle_stripe(struct stripe_head *sh)
 
 finish:
 	/* wait for this device to become unblocked */
-	if (conf->mddev->external && unlikely(s.blocked_rdev))
-		md_wait_for_blocked_rdev(s.blocked_rdev, conf->mddev);
+	if (unlikely(s.blocked_rdev)) {
+		if (conf->mddev->external)
+			md_wait_for_blocked_rdev(s.blocked_rdev,
+						 conf->mddev);
+		else
+			/* Internal metadata will immediately
+			 * be written by raid5d, so we don't
+			 * need to wait here.
+			 */
+			rdev_dec_pending(s.blocked_rdev,
+					 conf->mddev);
+	}
 
 	if (s.handle_bad_blocks)
 		for (i = disks; i--; ) {
@@ -3881,8 +3906,6 @@ static int chunk_aligned_read(struct mddev *mddev, struct bio * raid_bio)
 		raid_bio->bi_next = (void*)rdev;
 		align_bi->bi_bdev =  rdev->bdev;
 		align_bi->bi_flags &= ~(1 << BIO_SEG_VALID);
-		/* No reshape active, so we can trust rdev->data_offset */
-		align_bi->bi_sector += rdev->data_offset;
 
 		if (!bio_fits_rdev(align_bi) ||
 		    is_badblock(rdev, align_bi->bi_sector, align_bi->bi_size>>9,
@@ -3893,6 +3916,9 @@ static int chunk_aligned_read(struct mddev *mddev, struct bio * raid_bio)
 			return 0;
 		}
 
+		/* No reshape active, so we can trust rdev->data_offset */
+		align_bi->bi_sector += rdev->data_offset;
+
 		spin_lock_irq(&conf->device_lock);
 		wait_event_lock_irq(conf->wait_for_stripe,
 				    conf->quiesce == 0,
@@ -3971,7 +3997,6 @@ static void make_request(struct mddev *mddev, struct bio * bi)
 	struct stripe_head *sh;
 	const int rw = bio_data_dir(bi);
 	int remaining;
-	int plugged;
 
 	if (unlikely(bi->bi_rw & REQ_FLUSH)) {
 		md_flush_request(mddev, bi);
@@ -3990,7 +4015,6 @@ static void make_request(struct mddev *mddev, struct bio * bi)
 	bi->bi_next = NULL;
 	bi->bi_phys_segments = 1;	/* over-loaded to count active stripes */
 
-	plugged = mddev_check_plugged(mddev);
 	for (;logical_sector < last_sector; logical_sector += STRIPE_SECTORS) {
 		DEFINE_WAIT(w);
 		int previous;
@@ -4092,6 +4116,7 @@ static void make_request(struct mddev *mddev, struct bio * bi)
 			if ((bi->bi_rw & REQ_SYNC) &&
 			    !test_and_set_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
 				atomic_inc(&conf->preread_active_stripes);
+			mddev_check_plugged(mddev);
 			release_stripe(sh);
 		} else {
 			/* cannot get stripe for read-ahead, just give-up */
@@ -4099,10 +4124,7 @@ static void make_request(struct mddev *mddev, struct bio * bi)
 			finish_wait(&conf->wait_for_overlap, &w);
 			break;
 		}
-			
 	}
-	if (!plugged)
-		md_wakeup_thread(mddev->thread);
 
 	spin_lock_irq(&conf->device_lock);
 	remaining = raid5_dec_bi_phys_segments(bi);
@@ -4823,6 +4845,7 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	int raid_disk, memory, max_disks;
 	struct md_rdev *rdev;
 	struct disk_info *disk;
+	char pers_name[6];
 
 	if (mddev->new_level != 5
 	    && mddev->new_level != 4
@@ -4946,7 +4969,8 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 		printk(KERN_INFO "md/raid:%s: allocated %dkB\n",
 		       mdname(mddev), memory);
 
-	conf->thread = md_register_thread(raid5d, mddev, NULL);
+	sprintf(pers_name, "raid%d", mddev->new_level);
+	conf->thread = md_register_thread(raid5d, mddev, pers_name);
 	if (!conf->thread) {
 		printk(KERN_ERR
 		       "md/raid:%s: couldn't allocate thread.\n",
@@ -5465,10 +5489,9 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 	if (rdev->saved_raid_disk >= 0 &&
 	    rdev->saved_raid_disk >= first &&
 	    conf->disks[rdev->saved_raid_disk].rdev == NULL)
-		disk = rdev->saved_raid_disk;
-	else
-		disk = first;
-	for ( ; disk <= last ; disk++) {
+		first = rdev->saved_raid_disk;
+
+	for (disk = first; disk <= last; disk++) {
 		p = conf->disks + disk;
 		if (p->rdev == NULL) {
 			clear_bit(In_sync, &rdev->flags);
@@ -5477,8 +5500,11 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 			if (rdev->saved_raid_disk != disk)
 				conf->fullsync = 1;
 			rcu_assign_pointer(p->rdev, rdev);
-			break;
+			goto out;
 		}
+	}
+	for (disk = first; disk <= last; disk++) {
+		p = conf->disks + disk;
 		if (test_bit(WantReplacement, &p->rdev->flags) &&
 		    p->replacement == NULL) {
 			clear_bit(In_sync, &rdev->flags);
@@ -5490,6 +5516,7 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 			break;
 		}
 	}
+out:
 	print_raid5_conf(conf);
 	return err;
 }