btrfs: use a dedicated data structure for chunk maps

Currently we abuse the extent_map structure for two purposes:

1) To actually represent extents for inodes;
2) To represent chunk mappings.

This is odd and has several disadvantages:

1) To create a chunk map, we need to do two memory allocations: one for
   an extent_map structure and another one for a map_lookup structure, so
   more potential for an allocation failure and more complicated code to
   manage and link two structures;

2) For a chunk map we actually only use 3 fields (24 bytes) of the
   respective extent map structure: the 'start' field to have the logical
   start address of the chunk, the 'len' field to have the chunk's size,
   and the 'orig_block_len' field to contain the chunk's stripe size.

   Besides wasting a memory, it's also odd and not intuitive at all to
   have the stripe size in a field named 'orig_block_len'.

   We are also using 'block_len' of the extent_map structure to contain
   the chunk size, so we have 2 fields for the same value, 'len' and
   'block_len', which is pointless;

3) When an extent map is associated to a chunk mapping, we set the bit
   EXTENT_FLAG_FS_MAPPING on its flags and then make its member named
   'map_lookup' point to the associated map_lookup structure. This means
   that for an extent map associated to an inode extent, we are not using
   this 'map_lookup' pointer, so wasting 8 bytes (on a 64 bits platform);

4) Extent maps associated to a chunk mapping are never merged or split so
   it's pointless to use the existing extent map infrastructure.

So add a dedicated data structure named 'btrfs_chunk_map' to represent
chunk mappings, this is basically the existing map_lookup structure with
some extra fields:

1) 'start' to contain the chunk logical address;
2) 'chunk_len' to contain the chunk's length;
3) 'stripe_size' for the stripe size;
4) 'rb_node' for insertion into a rb tree;
5) 'refs' for reference counting.

This way we do a single memory allocation for chunk mappings and we don't
waste memory for them with unused/unnecessary fields from an extent_map.

We also save 8 bytes from the extent_map structure by removing the
'map_lookup' pointer, so the size of struct extent_map is reduced from
144 bytes down to 136 bytes, and we can now have 30 extents map per 4K
page instead of 28.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

fs/btrfs/block-group.h

@@ -5,6 +5,8 @@
 
 #include "free-space-cache.h"
 
+struct btrfs_chunk_map;
+
 enum btrfs_disk_cache_state {
 	BTRFS_DC_WRITTEN,
 	BTRFS_DC_ERROR,
@@ -243,7 +245,7 @@ struct btrfs_block_group {
 	u64 zone_unusable;
 	u64 zone_capacity;
 	u64 meta_write_pointer;
-	struct map_lookup *physical_map;
+	struct btrfs_chunk_map *physical_map;
 	struct list_head active_bg_list;
 	struct work_struct zone_finish_work;
 	struct extent_buffer *last_eb;
@@ -297,7 +299,7 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
 				struct btrfs_fs_info *fs_info,
 				const u64 chunk_offset);
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
-			     u64 group_start, struct extent_map *em);
+			     struct btrfs_chunk_map *map);
 void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info);
 void btrfs_mark_bg_unused(struct btrfs_block_group *bg);
 void btrfs_reclaim_bgs_work(struct work_struct *work);

fs/btrfs/dev-replace.c

@@ -550,8 +550,7 @@ bool btrfs_finish_block_group_to_copy(struct btrfs_device *srcdev,
 				      u64 physical)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
-	struct extent_map *em;
-	struct map_lookup *map;
+	struct btrfs_chunk_map *map;
 	u64 chunk_offset = cache->start;
 	int num_extents, cur_extent;
 	int i;
@@ -567,9 +566,8 @@ bool btrfs_finish_block_group_to_copy(struct btrfs_device *srcdev,
 	}
 	spin_unlock(&cache->lock);
 
-	em = btrfs_get_chunk_map(fs_info, chunk_offset, 1);
-	ASSERT(!IS_ERR(em));
-	map = em->map_lookup;
+	map = btrfs_get_chunk_map(fs_info, chunk_offset, 1);
+	ASSERT(!IS_ERR(map));
 
 	num_extents = 0;
 	cur_extent = 0;
@@ -583,7 +581,7 @@ bool btrfs_finish_block_group_to_copy(struct btrfs_device *srcdev,
 			cur_extent = i;
 	}
 
-	free_extent_map(em);
+	btrfs_free_chunk_map(map);
 
 	if (num_extents > 1 && cur_extent < num_extents - 1) {
 		/*
@@ -812,25 +810,23 @@ static void btrfs_dev_replace_update_device_in_mapping_tree(
 						struct btrfs_device *srcdev,
 						struct btrfs_device *tgtdev)
 {
-	struct extent_map_tree *em_tree = &fs_info->mapping_tree;
-	struct extent_map *em;
-	struct map_lookup *map;
 	u64 start = 0;
 	int i;
 
-	write_lock(&em_tree->lock);
+	write_lock(&fs_info->mapping_tree_lock);
 	do {
-		em = lookup_extent_mapping(em_tree, start, (u64)-1);
-		if (!em)
+		struct btrfs_chunk_map *map;
+
+		map = btrfs_find_chunk_map_nolock(fs_info, start, U64_MAX);
+		if (!map)
 			break;
-		map = em->map_lookup;
 		for (i = 0; i < map->num_stripes; i++)
 			if (srcdev == map->stripes[i].dev)
 				map->stripes[i].dev = tgtdev;
-		start = em->start + em->len;
-		free_extent_map(em);
+		start = map->start + map->chunk_len;
+		btrfs_free_chunk_map(map);
 	} while (start);
-	write_unlock(&em_tree->lock);
+	write_unlock(&fs_info->mapping_tree_lock);
 }
 
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,

fs/btrfs/disk-io.c

@@ -2720,7 +2720,8 @@ void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
 	INIT_LIST_HEAD(&fs_info->allocated_ebs);
 	spin_lock_init(&fs_info->eb_leak_lock);
 #endif
-	extent_map_tree_init(&fs_info->mapping_tree);
+	fs_info->mapping_tree = RB_ROOT_CACHED;
+	rwlock_init(&fs_info->mapping_tree_lock);
 	btrfs_init_block_rsv(&fs_info->global_block_rsv,
 			     BTRFS_BLOCK_RSV_GLOBAL);
 	btrfs_init_block_rsv(&fs_info->trans_block_rsv, BTRFS_BLOCK_RSV_TRANS);
@@ -3604,7 +3605,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 	btrfs_stop_all_workers(fs_info);
 	btrfs_free_block_groups(fs_info);
 fail_alloc:
-	btrfs_mapping_tree_free(&fs_info->mapping_tree);
+	btrfs_mapping_tree_free(fs_info);
 
 	iput(fs_info->btree_inode);
 fail:
@@ -4387,7 +4388,7 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
 
 	iput(fs_info->btree_inode);
 
-	btrfs_mapping_tree_free(&fs_info->mapping_tree);
+	btrfs_mapping_tree_free(fs_info);
 	btrfs_close_devices(fs_info->fs_devices);
 }
 

fs/btrfs/extent_map.c

@@ -67,8 +67,6 @@ void free_extent_map(struct extent_map *em)
 	if (refcount_dec_and_test(&em->refs)) {
 		WARN_ON(extent_map_in_tree(em));
 		WARN_ON(!list_empty(&em->list));
-		if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
-			kfree(em->map_lookup);
 		kmem_cache_free(extent_map_cache, em);
 	}
 }
@@ -217,13 +215,8 @@ static int mergable_maps(struct extent_map *prev, struct extent_map *next)
 	ASSERT(next->block_start != EXTENT_MAP_DELALLOC &&
 	       prev->block_start != EXTENT_MAP_DELALLOC);
 
-	if (prev->map_lookup || next->map_lookup)
-		ASSERT(test_bit(EXTENT_FLAG_FS_MAPPING, &prev->flags) &&
-		       test_bit(EXTENT_FLAG_FS_MAPPING, &next->flags));
-
 	if (extent_map_end(prev) == next->start &&
 	    prev->flags == next->flags &&
-	    prev->map_lookup == next->map_lookup &&
 	    ((next->block_start == EXTENT_MAP_HOLE &&
 	      prev->block_start == EXTENT_MAP_HOLE) ||
 	     (next->block_start == EXTENT_MAP_INLINE &&
@@ -361,39 +354,6 @@ static inline void setup_extent_mapping(struct extent_map_tree *tree,
 		try_merge_map(tree, em);
 }
 
-static void extent_map_device_set_bits(struct extent_map *em, unsigned bits)
-{
-	struct map_lookup *map = em->map_lookup;
-	u64 stripe_size = em->orig_block_len;
-	int i;
-
-	for (i = 0; i < map->num_stripes; i++) {
-		struct btrfs_io_stripe *stripe = &map->stripes[i];
-		struct btrfs_device *device = stripe->dev;
-
-		set_extent_bit(&device->alloc_state, stripe->physical,
-			       stripe->physical + stripe_size - 1,
-			       bits | EXTENT_NOWAIT, NULL);
-	}
-}
-
-static void extent_map_device_clear_bits(struct extent_map *em, unsigned bits)
-{
-	struct map_lookup *map = em->map_lookup;
-	u64 stripe_size = em->orig_block_len;
-	int i;
-
-	for (i = 0; i < map->num_stripes; i++) {
-		struct btrfs_io_stripe *stripe = &map->stripes[i];
-		struct btrfs_device *device = stripe->dev;
-
-		__clear_extent_bit(&device->alloc_state, stripe->physical,
-				   stripe->physical + stripe_size - 1,
-				   bits | EXTENT_NOWAIT,
-				   NULL, NULL);
-	}
-}
-
 /*
  * Add new extent map to the extent tree
  *
@@ -419,10 +379,6 @@ int add_extent_mapping(struct extent_map_tree *tree,
 		goto out;
 
 	setup_extent_mapping(tree, em, modified);
-	if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags)) {
-		extent_map_device_set_bits(em, CHUNK_ALLOCATED);
-		extent_map_device_clear_bits(em, CHUNK_TRIMMED);
-	}
 out:
 	return ret;
 }
@@ -506,8 +462,6 @@ void remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em)
 	rb_erase_cached(&em->rb_node, &tree->map);
 	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
 		list_del_init(&em->list);
-	if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
-		extent_map_device_clear_bits(em, CHUNK_ALLOCATED);
 	RB_CLEAR_NODE(&em->rb_node);
 }
 

fs/btrfs/extent_map.h

@@ -23,8 +23,6 @@ enum {
 	EXTENT_FLAG_LOGGING,
 	/* Filling in a preallocated extent */
 	EXTENT_FLAG_FILLING,
-	/* filesystem extent mapping type */
-	EXTENT_FLAG_FS_MAPPING,
 	/* This em is merged from two or more physically adjacent ems */
 	EXTENT_FLAG_MERGED,
 };
@@ -50,8 +48,6 @@ struct extent_map {
 	 */
 	u64 generation;
 	unsigned long flags;
-	/* Used for chunk mappings, flag EXTENT_FLAG_FS_MAPPING must be set */
-	struct map_lookup *map_lookup;
 	refcount_t refs;
 	unsigned int compress_type;
 	struct list_head list;

fs/btrfs/fs.h

@@ -398,7 +398,8 @@ struct btrfs_fs_info {
 	struct extent_io_tree excluded_extents;
 
 	/* logical->physical extent mapping */
-	struct extent_map_tree mapping_tree;
+	struct rb_root_cached mapping_tree;
+	rwlock_t mapping_tree_lock;
 
 	/*
 	 * Block reservation for extent, checksum, root tree and delayed dir

fs/btrfs/inode.c

@@ -10565,6 +10565,7 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_state *cached_state = NULL;
 	struct extent_map *em = NULL;
+	struct btrfs_chunk_map *map = NULL;
 	struct btrfs_device *device = NULL;
 	struct btrfs_swap_info bsi = {
 		.lowest_ppage = (sector_t)-1ULL,
@@ -10704,13 +10705,13 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 			goto out;
 		}
 
-		em = btrfs_get_chunk_map(fs_info, logical_block_start, len);
-		if (IS_ERR(em)) {
-			ret = PTR_ERR(em);
+		map = btrfs_get_chunk_map(fs_info, logical_block_start, len);
+		if (IS_ERR(map)) {
+			ret = PTR_ERR(map);
 			goto out;
 		}
 
-		if (em->map_lookup->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
+		if (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
 			btrfs_warn(fs_info,
 				   "swapfile must have single data profile");
 			ret = -EINVAL;
@@ -10718,23 +10719,23 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 		}
 
 		if (device == NULL) {
-			device = em->map_lookup->stripes[0].dev;
+			device = map->stripes[0].dev;
 			ret = btrfs_add_swapfile_pin(inode, device, false);
 			if (ret == 1)
 				ret = 0;
 			else if (ret)
 				goto out;
-		} else if (device != em->map_lookup->stripes[0].dev) {
+		} else if (device != map->stripes[0].dev) {
 			btrfs_warn(fs_info, "swapfile must be on one device");
 			ret = -EINVAL;
 			goto out;
 		}
 
-		physical_block_start = (em->map_lookup->stripes[0].physical +
-					(logical_block_start - em->start));
-		len = min(len, em->len - (logical_block_start - em->start));
-		free_extent_map(em);
-		em = NULL;
+		physical_block_start = (map->stripes[0].physical +
+					(logical_block_start - map->start));
+		len = min(len, map->chunk_len - (logical_block_start - map->start));
+		btrfs_free_chunk_map(map);
+		map = NULL;
 
 		bg = btrfs_lookup_block_group(fs_info, logical_block_start);
 		if (!bg) {
@@ -10787,6 +10788,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 out:
 	if (!IS_ERR_OR_NULL(em))
 		free_extent_map(em);
+	if (!IS_ERR_OR_NULL(map))
+		btrfs_free_chunk_map(map);
 
 	unlock_extent(io_tree, 0, isize - 1, &cached_state);
 

fs/btrfs/raid56.h

@@ -164,7 +164,7 @@ struct raid56_bio_trace_info {
 	u8 stripe_nr;
 };
 
-static inline int nr_data_stripes(const struct map_lookup *map)
+static inline int nr_data_stripes(const struct btrfs_chunk_map *map)
 {
 	return map->num_stripes - btrfs_nr_parity_stripes(map->type);
 }

fs/btrfs/scrub.c

@@ -1279,7 +1279,7 @@ static void scrub_throttle_dev_io(struct scrub_ctx *sctx, struct btrfs_device *d
  * return 0 if it is a data stripe, 1 means parity stripe.
  */
 static int get_raid56_logic_offset(u64 physical, int num,
-				   struct map_lookup *map, u64 *offset,
+				   struct btrfs_chunk_map *map, u64 *offset,
 				   u64 *stripe_start)
 {
 	int i;
@@ -1894,7 +1894,7 @@ static int queue_scrub_stripe(struct scrub_ctx *sctx, struct btrfs_block_group *
 static int scrub_raid56_parity_stripe(struct scrub_ctx *sctx,
 				      struct btrfs_device *scrub_dev,
 				      struct btrfs_block_group *bg,
-				      struct map_lookup *map,
+				      struct btrfs_chunk_map *map,
 				      u64 full_stripe_start)
 {
 	DECLARE_COMPLETION_ONSTACK(io_done);
@@ -2063,7 +2063,7 @@ static int scrub_raid56_parity_stripe(struct scrub_ctx *sctx,
  */
 static int scrub_simple_mirror(struct scrub_ctx *sctx,
 			       struct btrfs_block_group *bg,
-			       struct map_lookup *map,
+			       struct btrfs_chunk_map *map,
 			       u64 logical_start, u64 logical_length,
 			       struct btrfs_device *device,
 			       u64 physical, int mirror_num)
@@ -2124,7 +2124,7 @@ static int scrub_simple_mirror(struct scrub_ctx *sctx,
 }
 
 /* Calculate the full stripe length for simple stripe based profiles */
-static u64 simple_stripe_full_stripe_len(const struct map_lookup *map)
+static u64 simple_stripe_full_stripe_len(const struct btrfs_chunk_map *map)
 {
 	ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 |
 			    BTRFS_BLOCK_GROUP_RAID10));
@@ -2133,7 +2133,7 @@ static u64 simple_stripe_full_stripe_len(const struct map_lookup *map)
 }
 
 /* Get the logical bytenr for the stripe */
-static u64 simple_stripe_get_logical(struct map_lookup *map,
+static u64 simple_stripe_get_logical(struct btrfs_chunk_map *map,
 				     struct btrfs_block_group *bg,
 				     int stripe_index)
 {
@@ -2150,7 +2150,7 @@ static u64 simple_stripe_get_logical(struct map_lookup *map,
 }
 
 /* Get the mirror number for the stripe */
-static int simple_stripe_mirror_num(struct map_lookup *map, int stripe_index)
+static int simple_stripe_mirror_num(struct btrfs_chunk_map *map, int stripe_index)
 {
 	ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 |
 			    BTRFS_BLOCK_GROUP_RAID10));
@@ -2162,7 +2162,7 @@ static int simple_stripe_mirror_num(struct map_lookup *map, int stripe_index)
 
 static int scrub_simple_stripe(struct scrub_ctx *sctx,
 			       struct btrfs_block_group *bg,
-			       struct map_lookup *map,
+			       struct btrfs_chunk_map *map,
 			       struct btrfs_device *device,
 			       int stripe_index)
 {
@@ -2195,18 +2195,17 @@ static int scrub_simple_stripe(struct scrub_ctx *sctx,
 
 static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 					   struct btrfs_block_group *bg,
-					   struct extent_map *em,
+					   struct btrfs_chunk_map *map,
 					   struct btrfs_device *scrub_dev,
 					   int stripe_index)
 {
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
-	struct map_lookup *map = em->map_lookup;
 	const u64 profile = map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK;
 	const u64 chunk_logical = bg->start;
 	int ret;
 	int ret2;
 	u64 physical = map->stripes[stripe_index].physical;
-	const u64 dev_stripe_len = btrfs_calc_stripe_length(em);
+	const u64 dev_stripe_len = btrfs_calc_stripe_length(map);
 	const u64 physical_end = physical + dev_stripe_len;
 	u64 logical;
 	u64 logic_end;
@@ -2369,17 +2368,12 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 					  u64 dev_extent_len)
 {
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
-	struct extent_map_tree *map_tree = &fs_info->mapping_tree;
-	struct map_lookup *map;
-	struct extent_map *em;
+	struct btrfs_chunk_map *map;
 	int i;
 	int ret = 0;
 
-	read_lock(&map_tree->lock);
-	em = lookup_extent_mapping(map_tree, bg->start, bg->length);
-	read_unlock(&map_tree->lock);
-
-	if (!em) {
+	map = btrfs_find_chunk_map(fs_info, bg->start, bg->length);
+	if (!map) {
 		/*
 		 * Might have been an unused block group deleted by the cleaner
 		 * kthread or relocation.
@@ -2391,22 +2385,21 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 
 		return ret;
 	}
-	if (em->start != bg->start)
+	if (map->start != bg->start)
 		goto out;
-	if (em->len < dev_extent_len)
+	if (map->chunk_len < dev_extent_len)
 		goto out;
 
-	map = em->map_lookup;
 	for (i = 0; i < map->num_stripes; ++i) {
 		if (map->stripes[i].dev->bdev == scrub_dev->bdev &&
 		    map->stripes[i].physical == dev_offset) {
-			ret = scrub_stripe(sctx, bg, em, scrub_dev, i);
+			ret = scrub_stripe(sctx, bg, map, scrub_dev, i);
 			if (ret)
 				goto out;
 		}
 	}
 out:
-	free_extent_map(em);
+	btrfs_free_chunk_map(map);
 
 	return ret;
 }

fs/btrfs/tests/btrfs-tests.c

@@ -28,6 +28,7 @@ const char *test_error[] = {
 	[TEST_ALLOC_INODE]	     = "cannot allocate inode",
 	[TEST_ALLOC_BLOCK_GROUP]     = "cannot allocate block group",
 	[TEST_ALLOC_EXTENT_MAP]      = "cannot allocate extent map",
+	[TEST_ALLOC_CHUNK_MAP]       = "cannot allocate chunk map",
 };
 
 static const struct super_operations btrfs_test_super_ops = {
@@ -185,7 +186,7 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
 	}
 	spin_unlock(&fs_info->buffer_lock);
 
-	btrfs_mapping_tree_free(&fs_info->mapping_tree);
+	btrfs_mapping_tree_free(fs_info);
 	list_for_each_entry_safe(dev, tmp, &fs_info->fs_devices->devices,
 				 dev_list) {
 		btrfs_free_dummy_device(dev);

fs/btrfs/tests/btrfs-tests.h

@@ -23,6 +23,7 @@ enum {
 	TEST_ALLOC_INODE,
 	TEST_ALLOC_BLOCK_GROUP,
 	TEST_ALLOC_EXTENT_MAP,
+	TEST_ALLOC_CHUNK_MAP,
 };
 
 extern const char *test_error[];

fs/btrfs/tests/extent-map-tests.c

@@ -859,33 +859,21 @@ struct rmap_test_vector {
 static int test_rmap_block(struct btrfs_fs_info *fs_info,
 			   struct rmap_test_vector *test)
 {
-	struct extent_map *em;
-	struct map_lookup *map = NULL;
+	struct btrfs_chunk_map *map;
 	u64 *logical = NULL;
 	int i, out_ndaddrs, out_stripe_len;
 	int ret;
 
-	em = alloc_extent_map();
-	if (!em) {
-		test_std_err(TEST_ALLOC_EXTENT_MAP);
-		return -ENOMEM;
-	}
-
-	map = kmalloc(map_lookup_size(test->num_stripes), GFP_KERNEL);
+	map = btrfs_alloc_chunk_map(test->num_stripes, GFP_KERNEL);
 	if (!map) {
-		kfree(em);
-		test_std_err(TEST_ALLOC_EXTENT_MAP);
+		test_std_err(TEST_ALLOC_CHUNK_MAP);
 		return -ENOMEM;
 	}
 
-	set_bit(EXTENT_FLAG_FS_MAPPING, &em->flags);
 	/* Start at 4GiB logical address */
-	em->start = SZ_4G;
-	em->len = test->data_stripe_size * test->num_data_stripes;
-	em->block_len = em->len;
-	em->orig_block_len = test->data_stripe_size;
-	em->map_lookup = map;
-
+	map->start = SZ_4G;
+	map->chunk_len = test->data_stripe_size * test->num_data_stripes;
+	map->stripe_size = test->data_stripe_size;
 	map->num_stripes = test->num_stripes;
 	map->type = test->raid_type;
 
@@ -901,15 +889,13 @@ static int test_rmap_block(struct btrfs_fs_info *fs_info,
 		map->stripes[i].physical = test->data_stripe_phys_start[i];
 	}
 
-	write_lock(&fs_info->mapping_tree.lock);
-	ret = add_extent_mapping(&fs_info->mapping_tree, em, 0);
-	write_unlock(&fs_info->mapping_tree.lock);
+	ret = btrfs_add_chunk_map(fs_info, map);
 	if (ret) {
-		test_err("error adding block group mapping to mapping tree");
+		test_err("error adding chunk map to mapping tree");
 		goto out_free;
 	}
 
-	ret = btrfs_rmap_block(fs_info, em->start, btrfs_sb_offset(1),
+	ret = btrfs_rmap_block(fs_info, map->start, btrfs_sb_offset(1),
 			       &logical, &out_ndaddrs, &out_stripe_len);
 	if (ret || (out_ndaddrs == 0 && test->expected_mapped_addr)) {
 		test_err("didn't rmap anything but expected %d",
@@ -938,14 +924,8 @@ static int test_rmap_block(struct btrfs_fs_info *fs_info,
 
 	ret = 0;
 out:
-	write_lock(&fs_info->mapping_tree.lock);
-	remove_extent_mapping(&fs_info->mapping_tree, em);
-	write_unlock(&fs_info->mapping_tree.lock);
-	/* For us */
-	free_extent_map(em);
+	btrfs_remove_chunk_map(fs_info, map);
 out_free:
-	/* For the tree */
-	free_extent_map(em);
 	kfree(logical);
 	return ret;
 }

fs/btrfs/volumes.h

@@ -426,7 +426,8 @@ struct btrfs_discard_stripe {
 struct btrfs_io_context {
 	refcount_t refs;
 	struct btrfs_fs_info *fs_info;
-	u64 map_type; /* get from map_lookup->type */
+	/* Taken from struct btrfs_chunk_map::type. */
+	u64 map_type;
 	struct bio *orig_bio;
 	atomic_t error;
 	u16 max_errors;
@@ -529,18 +530,32 @@ struct btrfs_raid_attr {
 
 extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES];
 
-struct map_lookup {
+struct btrfs_chunk_map {
+	struct rb_node rb_node;
+	/* For mount time dev extent verification. */
+	int verified_stripes;
+	refcount_t refs;
+	u64 start;
+	u64 chunk_len;
+	u64 stripe_size;
 	u64 type;
 	int io_align;
 	int io_width;
 	int num_stripes;
 	int sub_stripes;
-	int verified_stripes; /* For mount time dev extent verification */
 	struct btrfs_io_stripe stripes[];
 };
 
-#define map_lookup_size(n) (sizeof(struct map_lookup) + \
-			    (sizeof(struct btrfs_io_stripe) * (n)))
+#define btrfs_chunk_map_size(n) (sizeof(struct btrfs_chunk_map) + \
+				 (sizeof(struct btrfs_io_stripe) * (n)))
+
+static inline void btrfs_free_chunk_map(struct btrfs_chunk_map *map)
+{
+	if (map && refcount_dec_and_test(&map->refs)) {
+		ASSERT(RB_EMPTY_NODE(&map->rb_node));
+		kfree(map);
+	}
+}
 
 struct btrfs_balance_args;
 struct btrfs_balance_progress;
@@ -624,7 +639,7 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
 int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
 struct btrfs_block_group *btrfs_create_chunk(struct btrfs_trans_handle *trans,
 					    u64 type);
-void btrfs_mapping_tree_free(struct extent_map_tree *tree);
+void btrfs_mapping_tree_free(struct btrfs_fs_info *fs_info);
 int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       blk_mode_t flags, void *holder);
 struct btrfs_device *btrfs_scan_one_device(const char *path, blk_mode_t flags,
@@ -680,13 +695,25 @@ int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
 			   u64 logical, u64 len);
 unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
 				    u64 logical);
-u64 btrfs_calc_stripe_length(const struct extent_map *em);
+u64 btrfs_calc_stripe_length(const struct btrfs_chunk_map *map);
 int btrfs_nr_parity_stripes(u64 type);
 int btrfs_chunk_alloc_add_chunk_item(struct btrfs_trans_handle *trans,
 				     struct btrfs_block_group *bg);
 int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset);
-struct extent_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info,
-				       u64 logical, u64 length);
+
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+struct btrfs_chunk_map *btrfs_alloc_chunk_map(int num_stripes, gfp_t gfp);
+int btrfs_add_chunk_map(struct btrfs_fs_info *fs_info, struct btrfs_chunk_map *map);
+#endif
+
+struct btrfs_chunk_map *btrfs_clone_chunk_map(struct btrfs_chunk_map *map, gfp_t gfp);
+struct btrfs_chunk_map *btrfs_find_chunk_map(struct btrfs_fs_info *fs_info,
+					     u64 logical, u64 length);
+struct btrfs_chunk_map *btrfs_find_chunk_map_nolock(struct btrfs_fs_info *fs_info,
+						    u64 logical, u64 length);
+struct btrfs_chunk_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info,
+					    u64 logical, u64 length);
+void btrfs_remove_chunk_map(struct btrfs_fs_info *fs_info, struct btrfs_chunk_map *map);
 void btrfs_release_disk_super(struct btrfs_super_block *super);
 
 static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,

fs/btrfs/zoned.c

@@ -1290,7 +1290,7 @@ struct zone_info {
 
 static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 				struct zone_info *info, unsigned long *active,
-				struct map_lookup *map)
+				struct btrfs_chunk_map *map)
 {
 	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
 	struct btrfs_device *device = map->stripes[zone_idx].dev;
@@ -1393,7 +1393,7 @@ static int btrfs_load_block_group_single(struct btrfs_block_group *bg,
 }
 
 static int btrfs_load_block_group_dup(struct btrfs_block_group *bg,
-				      struct map_lookup *map,
+				      struct btrfs_chunk_map *map,
 				      struct zone_info *zone_info,
 				      unsigned long *active)
 {
@@ -1435,7 +1435,7 @@ static int btrfs_load_block_group_dup(struct btrfs_block_group *bg,
 }
 
 static int btrfs_load_block_group_raid1(struct btrfs_block_group *bg,
-					struct map_lookup *map,
+					struct btrfs_chunk_map *map,
 					struct zone_info *zone_info,
 					unsigned long *active)
 {
@@ -1483,7 +1483,7 @@ static int btrfs_load_block_group_raid1(struct btrfs_block_group *bg,
 }
 
 static int btrfs_load_block_group_raid0(struct btrfs_block_group *bg,
-					struct map_lookup *map,
+					struct btrfs_chunk_map *map,
 					struct zone_info *zone_info,
 					unsigned long *active)
 {
@@ -1515,7 +1515,7 @@ static int btrfs_load_block_group_raid0(struct btrfs_block_group *bg,
 }
 
 static int btrfs_load_block_group_raid10(struct btrfs_block_group *bg,
-					 struct map_lookup *map,
+					 struct btrfs_chunk_map *map,
 					 struct zone_info *zone_info,
 					 unsigned long *active)
 {
@@ -1552,9 +1552,7 @@ static int btrfs_load_block_group_raid10(struct btrfs_block_group *bg,
 int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
-	struct extent_map_tree *em_tree = &fs_info->mapping_tree;
-	struct extent_map *em;
-	struct map_lookup *map;
+	struct btrfs_chunk_map *map;
 	u64 logical = cache->start;
 	u64 length = cache->length;
 	struct zone_info *zone_info = NULL;
@@ -1575,17 +1573,11 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 		return -EIO;
 	}
 
-	/* Get the chunk mapping */
-	read_lock(&em_tree->lock);
-	em = lookup_extent_mapping(em_tree, logical, length);
-	read_unlock(&em_tree->lock);
-
-	if (!em)
+	map = btrfs_find_chunk_map(fs_info, logical, length);
+	if (!map)
 		return -EINVAL;
 
-	map = em->map_lookup;
-
-	cache->physical_map = kmemdup(map, map_lookup_size(map->num_stripes), GFP_NOFS);
+	cache->physical_map = btrfs_clone_chunk_map(map, GFP_NOFS);
 	if (!cache->physical_map) {
 		ret = -ENOMEM;
 		goto out;
@@ -1687,12 +1679,11 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 			spin_unlock(&fs_info->zone_active_bgs_lock);
 		}
 	} else {
-		kfree(cache->physical_map);
+		btrfs_free_chunk_map(cache->physical_map);
 		cache->physical_map = NULL;
 	}
 	bitmap_free(active);
 	kfree(zone_info);
-	free_extent_map(em);
 
 	return ret;
 }
@@ -2082,7 +2073,7 @@ int btrfs_sync_zone_write_pointer(struct btrfs_device *tgt_dev, u64 logical,
 bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
-	struct map_lookup *map;
+	struct btrfs_chunk_map *map;
 	struct btrfs_device *device;
 	u64 physical;
 	const bool is_data = (block_group->flags & BTRFS_BLOCK_GROUP_DATA);
@@ -2194,7 +2185,7 @@ static void wait_eb_writebacks(struct btrfs_block_group *block_group)
 static int do_zone_finish(struct btrfs_block_group *block_group, bool fully_written)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
-	struct map_lookup *map;
+	struct btrfs_chunk_map *map;
 	const bool is_metadata = (block_group->flags &
 			(BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_SYSTEM));
 	int ret = 0;
@@ -2643,7 +2634,7 @@ void btrfs_check_active_zone_reservation(struct btrfs_fs_info *fs_info)
 	/* Release reservation for currently active block groups. */
 	spin_lock(&fs_info->zone_active_bgs_lock);
 	list_for_each_entry(block_group, &fs_info->zone_active_bgs, active_bg_list) {
-		struct map_lookup *map = block_group->physical_map;
+		struct btrfs_chunk_map *map = block_group->physical_map;
 
 		if (!(block_group->flags &
 		      (BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_SYSTEM)))

include/trace/events/btrfs.h

@@ -21,7 +21,7 @@ struct btrfs_delayed_data_ref;
 struct btrfs_delayed_ref_head;
 struct btrfs_block_group;
 struct btrfs_free_cluster;
-struct map_lookup;
+struct btrfs_chunk_map;
 struct extent_buffer;
 struct btrfs_work;
 struct btrfs_workqueue;
@@ -277,8 +277,7 @@ DEFINE_EVENT(btrfs__inode, btrfs_inode_evict,
 		{ (1 << EXTENT_FLAG_COMPRESSED), 	"COMPRESSED" 	},\
 		{ (1 << EXTENT_FLAG_PREALLOC), 		"PREALLOC" 	},\
 		{ (1 << EXTENT_FLAG_LOGGING),	 	"LOGGING" 	},\
-		{ (1 << EXTENT_FLAG_FILLING),	 	"FILLING" 	},\
-		{ (1 << EXTENT_FLAG_FS_MAPPING),	"FS_MAPPING"	})
+		{ (1 << EXTENT_FLAG_FILLING),		"FILLING"	})
 
 TRACE_EVENT_CONDITION(btrfs_get_extent,
 
@@ -1061,7 +1060,7 @@ DEFINE_EVENT(btrfs_delayed_ref_head,  run_delayed_ref_head,
 DECLARE_EVENT_CLASS(btrfs__chunk,
 
 	TP_PROTO(const struct btrfs_fs_info *fs_info,
-		 const struct map_lookup *map, u64 offset, u64 size),
+		 const struct btrfs_chunk_map *map, u64 offset, u64 size),
 
 	TP_ARGS(fs_info, map, offset, size),
 
@@ -1095,7 +1094,7 @@ DECLARE_EVENT_CLASS(btrfs__chunk,
 DEFINE_EVENT(btrfs__chunk,  btrfs_chunk_alloc,
 
 	TP_PROTO(const struct btrfs_fs_info *fs_info,
-		 const struct map_lookup *map, u64 offset, u64 size),
+		 const struct btrfs_chunk_map *map, u64 offset, u64 size),
 
 	TP_ARGS(fs_info, map, offset, size)
 );
@@ -1103,7 +1102,7 @@ DEFINE_EVENT(btrfs__chunk,  btrfs_chunk_alloc,
 DEFINE_EVENT(btrfs__chunk,  btrfs_chunk_free,
 
 	TP_PROTO(const struct btrfs_fs_info *fs_info,
-		 const struct map_lookup *map, u64 offset, u64 size),
+		 const struct btrfs_chunk_map *map, u64 offset, u64 size),
 
 	TP_ARGS(fs_info, map, offset, size)
 );