1 files changed, 47 insertions, 35 deletions

diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index 1b9795d75ef8..8abde6b8cedc 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -191,7 +191,6 @@ struct raid_dev {
 #define RT_FLAG_RS_BITMAP_LOADED        2
 #define RT_FLAG_UPDATE_SBS              3
 #define RT_FLAG_RESHAPE_RS              4
-#define RT_FLAG_KEEP_RS_FROZEN          5
 
 /* Array elements of 64 bit needed for rebuild/failed disk bits */
 #define DISKS_ARRAY_ELEMS ((MAX_RAID_DEVICES + (sizeof(uint64_t) * 8 - 1)) / sizeof(uint64_t) / 8)
@@ -861,6 +860,9 @@ static int validate_region_size(struct raid_set *rs, unsigned long region_size)
 {
         unsigned long min_region_size = rs->ti->len / (1 << 21);
 
+        if (rs_is_raid0(rs))
+                return 0;
+
         if (!region_size) {
                 /*
                  * Choose a reasonable default.  All figures in sectors.
@@ -930,6 +932,8 @@ static int validate_raid_redundancy(struct raid_set *rs)
                         rebuild_cnt++;
 
         switch (rs->raid_type->level) {
+        case 0:
+                break;
         case 1:
                 if (rebuild_cnt >= rs->md.raid_disks)
                         goto too_many;
@@ -2335,6 +2339,13 @@ static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
                 case 0:
                         break;
                 default:
+                        /*
+                         * We have to keep any raid0 data/metadata device pairs or
+                         * the MD raid0 personality will fail to start the array.
+                         */
+                        if (rs_is_raid0(rs))
+                                continue;
+
                         dev = container_of(rdev, struct raid_dev, rdev);
                         if (dev->meta_dev)
                                 dm_put_device(ti, dev->meta_dev);
@@ -2579,7 +2590,6 @@ static int rs_prepare_reshape(struct raid_set *rs)
                 } else {
                         /* Process raid1 without delta_disks */
                         mddev->raid_disks = rs->raid_disks;
-                        set_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags);
                         reshape = false;
                 }
         } else {
@@ -2590,7 +2600,6 @@ static int rs_prepare_reshape(struct raid_set *rs)
         if (reshape) {
                 set_bit(RT_FLAG_RESHAPE_RS, &rs->runtime_flags);
                 set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
-                set_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags);
         } else if (mddev->raid_disks < rs->raid_disks)
                 /* Create new superblocks and bitmaps, if any new disks */
                 set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
@@ -2902,7 +2911,6 @@ static int raid_ctr(struct dm_target *ti, unsigned int argc, char **argv)
                         goto bad;
 
                 set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
-                set_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags);
                 /* Takeover ain't recovery, so disable recovery */
                 rs_setup_recovery(rs, MaxSector);
                 rs_set_new(rs);
@@ -3386,21 +3394,28 @@ static void raid_postsuspend(struct dm_target *ti)
 {
         struct raid_set *rs = ti->private;
 
-        if (test_and_clear_bit(RT_FLAG_RS_RESUMED, &rs->runtime_flags)) {
-                if (!rs->md.suspended)
-                        mddev_suspend(&rs->md);
-                rs->md.ro = 1;
-        }
+        if (!rs->md.suspended)
+                mddev_suspend(&rs->md);
+
+        rs->md.ro = 1;
 }
 
 static void attempt_restore_of_faulty_devices(struct raid_set *rs)
 {
         int i;
-        uint64_t failed_devices, cleared_failed_devices = 0;
+        uint64_t cleared_failed_devices[DISKS_ARRAY_ELEMS];
         unsigned long flags;
+        bool cleared = false;
         struct dm_raid_superblock *sb;
+        struct mddev *mddev = &rs->md;
         struct md_rdev *r;
 
+        /* RAID personalities have to provide hot add/remove methods or we need to bail out. */
+        if (!mddev->pers || !mddev->pers->hot_add_disk || !mddev->pers->hot_remove_disk)
+                return;
+
+        memset(cleared_failed_devices, 0, sizeof(cleared_failed_devices));
+
         for (i = 0; i < rs->md.raid_disks; i++) {
                 r = &rs->dev[i].rdev;
                 if (test_bit(Faulty, &r->flags) && r->sb_page &&
@@ -3420,7 +3435,7 @@ static void attempt_restore_of_faulty_devices(struct raid_set *rs)
                          * ourselves.
                          */
                         if ((r->raid_disk >= 0) &&
-                            (r->mddev->pers->hot_remove_disk(r->mddev, r) != 0))
+                            (mddev->pers->hot_remove_disk(mddev, r) != 0))
                                 /* Failed to revive this device, try next */
                                 continue;
 
@@ -3430,22 +3445,30 @@ static void attempt_restore_of_faulty_devices(struct raid_set *rs)
                         clear_bit(Faulty, &r->flags);
                         clear_bit(WriteErrorSeen, &r->flags);
                         clear_bit(In_sync, &r->flags);
-                        if (r->mddev->pers->hot_add_disk(r->mddev, r)) {
+                        if (mddev->pers->hot_add_disk(mddev, r)) {
                                 r->raid_disk = -1;
                                 r->saved_raid_disk = -1;
                                 r->flags = flags;
                         } else {
                                 r->recovery_offset = 0;
-                                cleared_failed_devices |= 1 << i;
+                                set_bit(i, (void *) cleared_failed_devices);
+                                cleared = true;
                         }
                 }
         }
-        if (cleared_failed_devices) {
+
+        /* If any failed devices could be cleared, update all sbs failed_devices bits */
+        if (cleared) {
+                uint64_t failed_devices[DISKS_ARRAY_ELEMS];
+
                 rdev_for_each(r, &rs->md) {
                         sb = page_address(r->sb_page);
-                        failed_devices = le64_to_cpu(sb->failed_devices);
-                        failed_devices &= ~cleared_failed_devices;
-                        sb->failed_devices = cpu_to_le64(failed_devices);
+                        sb_retrieve_failed_devices(sb, failed_devices);
+
+                        for (i = 0; i < DISKS_ARRAY_ELEMS; i++)
+                                failed_devices[i] &= ~cleared_failed_devices[i];
+
+                        sb_update_failed_devices(sb, failed_devices);
                 }
         }
 }
@@ -3610,26 +3633,15 @@ static void raid_resume(struct dm_target *ti)
                  * devices are reachable again.
                  */
                 attempt_restore_of_faulty_devices(rs);
-        } else {
-                mddev->ro = 0;
-                mddev->in_sync = 0;
+        }
 
-                /*
-                 * When passing in flags to the ctr, we expect userspace
-                 * to reset them because they made it to the superblocks
-                 * and reload the mapping anyway.
-                 *
-                 * -> only unfreeze recovery in case of a table reload or
-                 *    we'll have a bogus recovery/reshape position
-                 *    retrieved from the superblock by the ctr because
-                 *    the ongoing recovery/reshape will change it after read.
-                 */
-                if (!test_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags))
-                        clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+        mddev->ro = 0;
+        mddev->in_sync = 0;
 
-                if (mddev->suspended)
-                        mddev_resume(mddev);
-        }
+        clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+
+        if (mddev->suspended)
+                mddev_resume(mddev);
 }
 
 static struct target_type raid_target = {