11 files changed, 185 insertions, 87 deletions

diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index d039de8322f0..b58b7a33914a 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -1084,6 +1084,7 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv)
         ti->split_io = dm_rh_get_region_size(ms->rh);
         ti->num_flush_requests = 1;
         ti->num_discard_requests = 1;
+        ti->discard_zeroes_data_unsupported = 1;
 
         ms->kmirrord_wq = alloc_workqueue("kmirrord",
                                           WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
@@ -1214,7 +1215,7 @@ static int mirror_end_io(struct dm_target *ti, struct bio *bio,
          * We need to dec pending if this was a write.
          */
         if (rw == WRITE) {
-                if (!(bio->bi_rw & REQ_FLUSH))
+                if (!(bio->bi_rw & (REQ_FLUSH | REQ_DISCARD)))
                         dm_rh_dec(ms->rh, map_context->ll);
                 return error;
         }
diff --git a/drivers/md/dm-region-hash.c b/drivers/md/dm-region-hash.c
index 7771ed212182..69732e03eb34 100644
--- a/drivers/md/dm-region-hash.c
+++ b/drivers/md/dm-region-hash.c
@@ -404,6 +404,9 @@ void dm_rh_mark_nosync(struct dm_region_hash *rh, struct bio *bio)
                 return;
         }
 
+        if (bio->bi_rw & REQ_DISCARD)
+                return;
+
         /* We must inform the log that the sync count has changed. */
         log->type->set_region_sync(log, region, 0);
 
@@ -524,7 +527,7 @@ void dm_rh_inc_pending(struct dm_region_hash *rh, struct bio_list *bios)
         struct bio *bio;
 
         for (bio = bios->head; bio; bio = bio->bi_next) {
-                if (bio->bi_rw & REQ_FLUSH)
+                if (bio->bi_rw & (REQ_FLUSH | REQ_DISCARD))
                         continue;
                 rh_inc(rh, dm_rh_bio_to_region(rh, bio));
         }
diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c
index 37fdaf81bd1f..68694da0d21d 100644
--- a/drivers/md/dm-thin.c
+++ b/drivers/md/dm-thin.c
@@ -1245,7 +1245,10 @@ static void process_discard(struct thin_c *tc, struct bio *bio)
 
                         cell_release_singleton(cell, bio);
                         cell_release_singleton(cell2, bio);
-                        remap_and_issue(tc, bio, lookup_result.block);
+                        if ((!lookup_result.shared) && pool->pf.discard_passdown)
+                                remap_and_issue(tc, bio, lookup_result.block);
+                        else
+                                bio_endio(bio, 0);
                 }
                 break;
 
@@ -2292,6 +2295,13 @@ static int process_reserve_metadata_snap_mesg(unsigned argc, char **argv, struct
         if (r)
                 return r;
 
+        r = dm_pool_commit_metadata(pool->pmd);
+        if (r) {
+                DMERR("%s: dm_pool_commit_metadata() failed, error = %d",
+                      __func__, r);
+                return r;
+        }
+
         r = dm_pool_reserve_metadata_snap(pool->pmd);
         if (r)
                 DMWARN("reserve_metadata_snap message failed.");
@@ -2621,6 +2631,7 @@ static int thin_ctr(struct dm_target *ti, unsigned argc, char **argv)
         if (tc->pool->pf.discard_enabled) {
                 ti->discards_supported = 1;
                 ti->num_discard_requests = 1;
+                ti->discard_zeroes_data_unsupported = 1;
         }
 
         dm_put(pool_md);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 1c2f9048e1ae..d5ab4493c8be 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2931,6 +2931,7 @@ offset_store(struct md_rdev *rdev, const char *buf, size_t len)
                  * can be sane */
                 return -EBUSY;
         rdev->data_offset = offset;
+        rdev->new_data_offset = offset;
         return len;
 }
 
@@ -3926,8 +3927,8 @@ array_state_show(struct mddev *mddev, char *page)
         return sprintf(page, "%s\n", array_states[st]);
 }
 
-static int do_md_stop(struct mddev * mddev, int ro, int is_open);
-static int md_set_readonly(struct mddev * mddev, int is_open);
+static int do_md_stop(struct mddev * mddev, int ro, struct block_device *bdev);
+static int md_set_readonly(struct mddev * mddev, struct block_device *bdev);
 static int do_md_run(struct mddev * mddev);
 static int restart_array(struct mddev *mddev);
 
@@ -3943,14 +3944,14 @@ array_state_store(struct mddev *mddev, const char *buf, size_t len)
                 /* stopping an active array */
                 if (atomic_read(&mddev->openers) > 0)
                         return -EBUSY;
-                err = do_md_stop(mddev, 0, 0);
+                err = do_md_stop(mddev, 0, NULL);
                 break;
         case inactive:
                 /* stopping an active array */
                 if (mddev->pers) {
                         if (atomic_read(&mddev->openers) > 0)
                                 return -EBUSY;
-                        err = do_md_stop(mddev, 2, 0);
+                        err = do_md_stop(mddev, 2, NULL);
                 } else
                         err = 0; /* already inactive */
                 break;
@@ -3958,7 +3959,7 @@ array_state_store(struct mddev *mddev, const char *buf, size_t len)
                 break; /* not supported yet */
         case readonly:
                 if (mddev->pers)
-                        err = md_set_readonly(mddev, 0);
+                        err = md_set_readonly(mddev, NULL);
                 else {
                         mddev->ro = 1;
                         set_disk_ro(mddev->gendisk, 1);
@@ -3968,7 +3969,7 @@ array_state_store(struct mddev *mddev, const char *buf, size_t len)
         case read_auto:
                 if (mddev->pers) {
                         if (mddev->ro == 0)
-                                err = md_set_readonly(mddev, 0);
+                                err = md_set_readonly(mddev, NULL);
                         else if (mddev->ro == 1)
                                 err = restart_array(mddev);
                         if (err == 0) {
@@ -5351,15 +5352,17 @@ void md_stop(struct mddev *mddev)
 }
 EXPORT_SYMBOL_GPL(md_stop);
 
-static int md_set_readonly(struct mddev *mddev, int is_open)
+static int md_set_readonly(struct mddev *mddev, struct block_device *bdev)
 {
         int err = 0;
         mutex_lock(&mddev->open_mutex);
-        if (atomic_read(&mddev->openers) > is_open) {
+        if (atomic_read(&mddev->openers) > !!bdev) {
                 printk("md: %s still in use.\n",mdname(mddev));
                 err = -EBUSY;
                 goto out;
         }
+        if (bdev)
+                sync_blockdev(bdev);
         if (mddev->pers) {
                 __md_stop_writes(mddev);
 
@@ -5381,18 +5384,26 @@ out:
  *   0 - completely stop and dis-assemble array
  *   2 - stop but do not disassemble array
  */
-static int do_md_stop(struct mddev * mddev, int mode, int is_open)
+static int do_md_stop(struct mddev * mddev, int mode,
+                      struct block_device *bdev)
 {
         struct gendisk *disk = mddev->gendisk;
         struct md_rdev *rdev;
 
         mutex_lock(&mddev->open_mutex);
-        if (atomic_read(&mddev->openers) > is_open ||
+        if (atomic_read(&mddev->openers) > !!bdev ||
             mddev->sysfs_active) {
                 printk("md: %s still in use.\n",mdname(mddev));
                 mutex_unlock(&mddev->open_mutex);
                 return -EBUSY;
         }
+        if (bdev)
+                /* It is possible IO was issued on some other
+                 * open file which was closed before we took ->open_mutex.
+                 * As that was not the last close __blkdev_put will not
+                 * have called sync_blockdev, so we must.
+                 */
+                sync_blockdev(bdev);
 
         if (mddev->pers) {
                 if (mddev->ro)
@@ -5466,7 +5477,7 @@ static void autorun_array(struct mddev *mddev)
         err = do_md_run(mddev);
         if (err) {
                 printk(KERN_WARNING "md: do_md_run() returned %d\n", err);
-                do_md_stop(mddev, 0, 0);
+                do_md_stop(mddev, 0, NULL);
         }
 }
 
@@ -5784,8 +5795,7 @@ 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.
                          */
@@ -6482,11 +6492,11 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
                         goto done_unlock;
 
                 case STOP_ARRAY:
-                        err = do_md_stop(mddev, 0, 1);
+                        err = do_md_stop(mddev, 0, bdev);
                         goto done_unlock;
 
                 case STOP_ARRAY_RO:
-                        err = md_set_readonly(mddev, 1);
+                        err = md_set_readonly(mddev, bdev);
                         goto done_unlock;
 
                 case BLKROSET:
@@ -6751,7 +6761,7 @@ 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 +7308,7 @@ 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 +7458,7 @@ 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 +7564,7 @@ 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/persistent-data/dm-space-map-checker.c b/drivers/md/persistent-data/dm-space-map-checker.c
index 50ed53bf4aa2..fc90c11620ad 100644
--- a/drivers/md/persistent-data/dm-space-map-checker.c
+++ b/drivers/md/persistent-data/dm-space-map-checker.c
@@ -8,6 +8,7 @@
 
 #include <linux/device-mapper.h>
 #include <linux/export.h>
+#include <linux/vmalloc.h>
 
 #ifdef CONFIG_DM_DEBUG_SPACE_MAPS
 
@@ -89,13 +90,23 @@ static int ca_create(struct count_array *ca, struct dm_space_map *sm)
 
         ca->nr = nr_blocks;
         ca->nr_free = nr_blocks;
-        ca->counts = kzalloc(sizeof(*ca->counts) * nr_blocks, GFP_KERNEL);
-        if (!ca->counts)
-                return -ENOMEM;
+
+        if (!nr_blocks)
+                ca->counts = NULL;
+        else {
+                ca->counts = vzalloc(sizeof(*ca->counts) * nr_blocks);
+                if (!ca->counts)
+                        return -ENOMEM;
+        }
 
         return 0;
 }
 
+static void ca_destroy(struct count_array *ca)
+{
+        vfree(ca->counts);
+}
+
 static int ca_load(struct count_array *ca, struct dm_space_map *sm)
 {
         int r;
@@ -126,12 +137,14 @@ static int ca_load(struct count_array *ca, struct dm_space_map *sm)
 static int ca_extend(struct count_array *ca, dm_block_t extra_blocks)
 {
         dm_block_t nr_blocks = ca->nr + extra_blocks;
-        uint32_t *counts = kzalloc(sizeof(*counts) * nr_blocks, GFP_KERNEL);
+        uint32_t *counts = vzalloc(sizeof(*counts) * nr_blocks);
         if (!counts)
                 return -ENOMEM;
 
-        memcpy(counts, ca->counts, sizeof(*counts) * ca->nr);
-        kfree(ca->counts);
+        if (ca->counts) {
+                memcpy(counts, ca->counts, sizeof(*counts) * ca->nr);
+                ca_destroy(ca);
+        }
         ca->nr = nr_blocks;
         ca->nr_free += extra_blocks;
         ca->counts = counts;
@@ -151,11 +164,6 @@ static int ca_commit(struct count_array *old, struct count_array *new)
         return 0;
 }
 
-static void ca_destroy(struct count_array *ca)
-{
-        kfree(ca->counts);
-}
-
 /*----------------------------------------------------------------*/
 
 struct sm_checker {
@@ -343,25 +351,25 @@ struct dm_space_map *dm_sm_checker_create(struct dm_space_map *sm)
         int r;
         struct sm_checker *smc;
 
-        if (!sm)
-                return NULL;
+        if (IS_ERR_OR_NULL(sm))
+                return ERR_PTR(-EINVAL);
 
         smc = kmalloc(sizeof(*smc), GFP_KERNEL);
         if (!smc)
-                return NULL;
+                return ERR_PTR(-ENOMEM);
 
         memcpy(&smc->sm, &ops_, sizeof(smc->sm));
         r = ca_create(&smc->old_counts, sm);
         if (r) {
                 kfree(smc);
-                return NULL;
+                return ERR_PTR(r);
         }
 
         r = ca_create(&smc->counts, sm);
         if (r) {
                 ca_destroy(&smc->old_counts);
                 kfree(smc);
-                return NULL;
+                return ERR_PTR(r);
         }
 
         smc->real_sm = sm;
@@ -371,7 +379,7 @@ struct dm_space_map *dm_sm_checker_create(struct dm_space_map *sm)
                 ca_destroy(&smc->counts);
                 ca_destroy(&smc->old_counts);
                 kfree(smc);
-                return NULL;
+                return ERR_PTR(r);
         }
 
         r = ca_commit(&smc->old_counts, &smc->counts);
@@ -379,7 +387,7 @@ struct dm_space_map *dm_sm_checker_create(struct dm_space_map *sm)
                 ca_destroy(&smc->counts);
                 ca_destroy(&smc->old_counts);
                 kfree(smc);
-                return NULL;
+                return ERR_PTR(r);
         }
 
         return &smc->sm;
@@ -391,25 +399,25 @@ struct dm_space_map *dm_sm_checker_create_fresh(struct dm_space_map *sm)
         int r;
         struct sm_checker *smc;
 
-        if (!sm)
-                return NULL;
+        if (IS_ERR_OR_NULL(sm))
+                return ERR_PTR(-EINVAL);
 
         smc = kmalloc(sizeof(*smc), GFP_KERNEL);
         if (!smc)
-                return NULL;
+                return ERR_PTR(-ENOMEM);
 
         memcpy(&smc->sm, &ops_, sizeof(smc->sm));
         r = ca_create(&smc->old_counts, sm);
         if (r) {
                 kfree(smc);
-                return NULL;
+                return ERR_PTR(r);
         }
 
         r = ca_create(&smc->counts, sm);
         if (r) {
                 ca_destroy(&smc->old_counts);
                 kfree(smc);
-                return NULL;
+                return ERR_PTR(r);
         }
 
         smc->real_sm = sm;
diff --git a/drivers/md/persistent-data/dm-space-map-disk.c b/drivers/md/persistent-data/dm-space-map-disk.c
index fc469ba9f627..3d0ed5332883 100644
--- a/drivers/md/persistent-data/dm-space-map-disk.c
+++ b/drivers/md/persistent-data/dm-space-map-disk.c
@@ -290,7 +290,16 @@ struct dm_space_map *dm_sm_disk_create(struct dm_transaction_manager *tm,
                                        dm_block_t nr_blocks)
 {
         struct dm_space_map *sm = dm_sm_disk_create_real(tm, nr_blocks);
-        return dm_sm_checker_create_fresh(sm);
+        struct dm_space_map *smc;
+
+        if (IS_ERR_OR_NULL(sm))
+                return sm;
+
+        smc = dm_sm_checker_create_fresh(sm);
+        if (IS_ERR(smc))
+                dm_sm_destroy(sm);
+
+        return smc;
 }
 EXPORT_SYMBOL_GPL(dm_sm_disk_create);
 
diff --git a/drivers/md/persistent-data/dm-transaction-manager.c b/drivers/md/persistent-data/dm-transaction-manager.c
index 400fe144c0cd..e5604b32d91f 100644
--- a/drivers/md/persistent-data/dm-transaction-manager.c
+++ b/drivers/md/persistent-data/dm-transaction-manager.c
@@ -138,6 +138,9 @@ EXPORT_SYMBOL_GPL(dm_tm_create_non_blocking_clone);
 
 void dm_tm_destroy(struct dm_transaction_manager *tm)
 {
+        if (!tm->is_clone)
+                wipe_shadow_table(tm);
+
         kfree(tm);
 }
 EXPORT_SYMBOL_GPL(dm_tm_destroy);
@@ -344,8 +347,10 @@ static int dm_tm_create_internal(struct dm_block_manager *bm,
                 }
 
                 *sm = dm_sm_checker_create(inner);
-                if (!*sm)
+                if (IS_ERR(*sm)) {
+                        r = PTR_ERR(*sm);
                         goto bad2;
+                }
 
         } else {
                 r = dm_bm_write_lock(dm_tm_get_bm(*tm), sb_location,
@@ -364,8 +369,10 @@ static int dm_tm_create_internal(struct dm_block_manager *bm,
                 }
 
                 *sm = dm_sm_checker_create(inner);
-                if (!*sm)
+                if (IS_ERR(*sm)) {
+                        r = PTR_ERR(*sm);
                         goto bad2;
+                }
         }
 
         return 0;
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 835de7168cd3..cacd008d6864 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)
@@ -1821,8 +1818,14 @@ static void sync_request_write(struct mddev *mddev, struct r1bio *r1_bio)
 
         if (atomic_dec_and_test(&r1_bio->remaining)) {
                 /* if we're here, all write(s) have completed, so clean up */
-                md_done_sync(mddev, r1_bio->sectors, 1);
-                put_buf(r1_bio);
+                int s = r1_bio->sectors;
+                if (test_bit(R1BIO_MadeGood, &r1_bio->state) ||
+                    test_bit(R1BIO_WriteError, &r1_bio->state))
+                        reschedule_retry(r1_bio);
+                else {
+                        put_buf(r1_bio);
+                        md_done_sync(mddev, s, 1);
+                }
         }
 }
 
@@ -2488,9 +2491,10 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
          */
         if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
                 atomic_set(&r1_bio->remaining, read_targets);
-                for (i = 0; i < conf->raid_disks * 2; i++) {
+                for (i = 0; i < conf->raid_disks * 2 && read_targets; i++) {
                         bio = r1_bio->bios[i];
                         if (bio->bi_end_io == end_sync_read) {
+                                read_targets--;
                                 md_sync_acct(bio->bi_bdev, nr_sectors);
                                 generic_make_request(bio);
                         }
@@ -2550,6 +2554,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
         err = -EINVAL;
         spin_lock_init(&conf->device_lock);
         rdev_for_each(rdev, mddev) {
+                struct request_queue *q;
                 int disk_idx = rdev->raid_disk;
                 if (disk_idx >= mddev->raid_disks
                     || disk_idx < 0)
@@ -2562,6 +2567,9 @@ static struct r1conf *setup_conf(struct mddev *mddev)
                 if (disk->rdev)
                         goto abort;
                 disk->rdev = rdev;
+                q = bdev_get_queue(rdev->bdev);
+                if (q->merge_bvec_fn)
+                        mddev->merge_check_needed = 1;
 
                 disk->head_position = 0;
         }
@@ -2617,7 +2625,7 @@ 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 987db37cb875..8da6282254c3 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))
+                        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;
 
@@ -3475,6 +3481,7 @@ static int run(struct mddev *mddev)
 
         rdev_for_each(rdev, mddev) {
                 long long diff;
+                struct request_queue *q;
 
                 disk_idx = rdev->raid_disk;
                 if (disk_idx < 0)
@@ -3493,6 +3500,9 @@ static int run(struct mddev *mddev)
                                 goto out_free_conf;
                         disk->rdev = rdev;
                 }
+                q = bdev_get_queue(rdev->bdev);
+                if (q->merge_bvec_fn)
+                        mddev->merge_check_needed = 1;
                 diff = (rdev->new_data_offset - rdev->data_offset);
                 if (!mddev->reshape_backwards)
                         diff = -diff;
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;
 }