4 files changed, 68 insertions, 15 deletions

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 7e65bad522cb..ac89a5deaca2 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -238,15 +238,47 @@ static struct page *read_sb_page(mddev_t *mddev, long offset, unsigned long inde
 
 }
 
+static mdk_rdev_t *next_active_rdev(mdk_rdev_t *rdev, mddev_t *mddev)
+{
+        /* Iterate the disks of an mddev, using rcu to protect access to the
+         * linked list, and raising the refcount of devices we return to ensure
+         * they don't disappear while in use.
+         * As devices are only added or removed when raid_disk is < 0 and
+         * nr_pending is 0 and In_sync is clear, the entries we return will
+         * still be in the same position on the list when we re-enter
+         * list_for_each_continue_rcu.
+         */
+        struct list_head *pos;
+        rcu_read_lock();
+        if (rdev == NULL)
+                /* start at the beginning */
+                pos = &mddev->disks;
+        else {
+                /* release the previous rdev and start from there. */
+                rdev_dec_pending(rdev, mddev);
+                pos = &rdev->same_set;
+        }
+        list_for_each_continue_rcu(pos, &mddev->disks) {
+                rdev = list_entry(pos, mdk_rdev_t, same_set);
+                if (rdev->raid_disk >= 0 &&
+                    test_bit(In_sync, &rdev->flags) &&
+                    !test_bit(Faulty, &rdev->flags)) {
+                        /* this is a usable devices */
+                        atomic_inc(&rdev->nr_pending);
+                        rcu_read_unlock();
+                        return rdev;
+                }
+        }
+        rcu_read_unlock();
+        return NULL;
+}
+
 static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 {
-        mdk_rdev_t *rdev;
+        mdk_rdev_t *rdev = NULL;
         mddev_t *mddev = bitmap->mddev;
 
-        rcu_read_lock();
-        rdev_for_each_rcu(rdev, mddev)
-                if (test_bit(In_sync, &rdev->flags)
-                    && !test_bit(Faulty, &rdev->flags)) {
+        while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
                         int size = PAGE_SIZE;
                         if (page->index == bitmap->file_pages-1)
                                 size = roundup(bitmap->last_page_size,
@@ -281,8 +313,7 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
                                        + page->index * (PAGE_SIZE/512),
                                        size,
                                        page);
-                }
-        rcu_read_unlock();
+        }
 
         if (wait)
                 md_super_wait(mddev);
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 71dd65aa31b6..c2fcf28b4c70 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -63,6 +63,7 @@ struct multipath {
 
         const char *hw_handler_name;
         struct work_struct activate_path;
+        struct pgpath *pgpath_to_activate;
         unsigned nr_priority_groups;
         struct list_head priority_groups;
         unsigned pg_init_required;      /* pg_init needs calling? */
@@ -146,6 +147,7 @@ static struct priority_group *alloc_priority_group(void)
 
 static void free_pgpaths(struct list_head *pgpaths, struct dm_target *ti)
 {
+        unsigned long flags;
         struct pgpath *pgpath, *tmp;
         struct multipath *m = ti->private;
 
@@ -154,6 +156,10 @@ static void free_pgpaths(struct list_head *pgpaths, struct dm_target *ti)
                 if (m->hw_handler_name)
                         scsi_dh_detach(bdev_get_queue(pgpath->path.dev->bdev));
                 dm_put_device(ti, pgpath->path.dev);
+                spin_lock_irqsave(&m->lock, flags);
+                if (m->pgpath_to_activate == pgpath)
+                        m->pgpath_to_activate = NULL;
+                spin_unlock_irqrestore(&m->lock, flags);
                 free_pgpath(pgpath);
         }
 }
@@ -421,6 +427,7 @@ static void process_queued_ios(struct work_struct *work)
                 __choose_pgpath(m);
 
         pgpath = m->current_pgpath;
+        m->pgpath_to_activate = m->current_pgpath;
 
         if ((pgpath && !m->queue_io) ||
             (!pgpath && !m->queue_if_no_path))
@@ -1093,8 +1100,15 @@ static void activate_path(struct work_struct *work)
         int ret;
         struct multipath *m =
                 container_of(work, struct multipath, activate_path);
-        struct dm_path *path = &m->current_pgpath->path;
+        struct dm_path *path;
+        unsigned long flags;
 
+        spin_lock_irqsave(&m->lock, flags);
+        path = &m->pgpath_to_activate->path;
+        m->pgpath_to_activate = NULL;
+        spin_unlock_irqrestore(&m->lock, flags);
+        if (!path)
+                return;
         ret = scsi_dh_activate(bdev_get_queue(path->dev->bdev));
         pg_init_done(path, ret);
 }
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index bca448e11878..ace998ce59f6 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -837,12 +837,14 @@ static int dm_merge_bvec(struct request_queue *q,
         struct dm_table *map = dm_get_table(md);
         struct dm_target *ti;
         sector_t max_sectors;
-        int max_size;
+        int max_size = 0;
 
         if (unlikely(!map))
-                return 0;
+                goto out;
 
         ti = dm_table_find_target(map, bvm->bi_sector);
+        if (!dm_target_is_valid(ti))
+                goto out_table;
 
         /*
          * Find maximum amount of I/O that won't need splitting
@@ -861,14 +863,16 @@ static int dm_merge_bvec(struct request_queue *q,
         if (max_size && ti->type->merge)
                 max_size = ti->type->merge(ti, bvm, biovec, max_size);
 
+out_table:
+        dm_table_put(map);
+
+out:
         /*
          * Always allow an entire first page
          */
         if (max_size <= biovec->bv_len && !(bvm->bi_size >> SECTOR_SHIFT))
                 max_size = biovec->bv_len;
 
-        dm_table_put(map);
-
         return max_size;
 }
 
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 8cfadc5bd2ba..deeac4b44173 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3841,8 +3841,6 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 
                 del_timer_sync(&mddev->safemode_timer);
 
-                invalidate_partition(disk, 0);
-
                 switch(mode) {
                 case 1: /* readonly */
                         err  = -ENXIO;
@@ -5763,7 +5761,11 @@ void md_do_sync(mddev_t *mddev)
                                          * time 'round when curr_resync == 2
                                          */
                                         continue;
-                                prepare_to_wait(&resync_wait, &wq, TASK_UNINTERRUPTIBLE);
+                                /* We need to wait 'interruptible' so as not to
+                                 * contribute to the load average, and not to
+                                 * be caught by 'softlockup'
+                                 */
+                                prepare_to_wait(&resync_wait, &wq, TASK_INTERRUPTIBLE);
                                 if (!kthread_should_stop() &&
                                     mddev2->curr_resync >= mddev->curr_resync) {
                                         printk(KERN_INFO "md: delaying %s of %s"
@@ -5771,6 +5773,8 @@ void md_do_sync(mddev_t *mddev)
                                                " share one or more physical units)\n",
                                                desc, mdname(mddev), mdname(mddev2));
                                         mddev_put(mddev2);
+                                        if (signal_pending(current))
+                                                flush_signals(current);
                                         schedule();
                                         finish_wait(&resync_wait, &wq);
                                         goto try_again;