btrfs: zoned: make zone finishing multi stripe capable
authorJohannes Thumshirn <johannes.thumshirn@wdc.com>
Wed, 26 Jan 2022 13:46:21 +0000 (05:46 -0800)
committerDavid Sterba <dsterba@suse.com>
Mon, 14 Mar 2022 12:13:49 +0000 (13:13 +0100)
Currently finishing of a zone only works if the block group isn't
spanning more than one zone.

This limitation is purely artificial and can be easily expanded to block
groups being places across multiple zones.

This is a preparation for allowing DUP and later more complex block-group
profiles on zoned btrfs.

Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
fs/btrfs/zoned.c

index e3d1984..987d245 100644 (file)
@@ -1845,19 +1845,12 @@ int btrfs_zone_finish(struct btrfs_block_group *block_group)
        struct btrfs_device *device;
        u64 physical;
        int ret = 0;
+       int i;
 
        if (!btrfs_is_zoned(fs_info))
                return 0;
 
        map = block_group->physical_map;
-       /* Currently support SINGLE profile only */
-       ASSERT(map->num_stripes == 1);
-
-       device = map->stripes[0].dev;
-       physical = map->stripes[0].physical;
-
-       if (device->zone_info->max_active_zones == 0)
-               return 0;
 
        spin_lock(&block_group->lock);
        if (!block_group->zone_is_active) {
@@ -1909,25 +1902,34 @@ int btrfs_zone_finish(struct btrfs_block_group *block_group)
        btrfs_clear_data_reloc_bg(block_group);
        spin_unlock(&block_group->lock);
 
-       ret = blkdev_zone_mgmt(device->bdev, REQ_OP_ZONE_FINISH,
-                              physical >> SECTOR_SHIFT,
-                              device->zone_info->zone_size >> SECTOR_SHIFT,
-                              GFP_NOFS);
-       btrfs_dec_block_group_ro(block_group);
+       for (i = 0; i < map->num_stripes; i++) {
+               device = map->stripes[i].dev;
+               physical = map->stripes[i].physical;
 
-       if (!ret) {
-               btrfs_dev_clear_active_zone(device, physical);
+               if (device->zone_info->max_active_zones == 0)
+                       continue;
 
-               spin_lock(&fs_info->zone_active_bgs_lock);
-               ASSERT(!list_empty(&block_group->active_bg_list));
-               list_del_init(&block_group->active_bg_list);
-               spin_unlock(&fs_info->zone_active_bgs_lock);
+               ret = blkdev_zone_mgmt(device->bdev, REQ_OP_ZONE_FINISH,
+                                      physical >> SECTOR_SHIFT,
+                                      device->zone_info->zone_size >> SECTOR_SHIFT,
+                                      GFP_NOFS);
+
+               if (ret)
+                       return ret;
 
-               /* For active_bg_list */
-               btrfs_put_block_group(block_group);
+               btrfs_dev_clear_active_zone(device, physical);
        }
+       btrfs_dec_block_group_ro(block_group);
 
-       return ret;
+       spin_lock(&fs_info->zone_active_bgs_lock);
+       ASSERT(!list_empty(&block_group->active_bg_list));
+       list_del_init(&block_group->active_bg_list);
+       spin_unlock(&fs_info->zone_active_bgs_lock);
+
+       /* For active_bg_list */
+       btrfs_put_block_group(block_group);
+
+       return 0;
 }
 
 bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices, u64 flags)