Merge tag 'for-5.7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "A number of core changes that make things work better in general, code
  is simpler and cleaner.

  Core changes:

   - per-inode file extent tree, for in memory tracking of contiguous
     extent ranges to make sure i_size adjustments are accurate

   - tree root structures are protected by reference counts, replacing
     SRCU that did not cover some cases

   - leak detector for tree root structures

   - per-transaction pinned extent tracking

   - buffer heads are replaced by bios for super block access

   - speedup of extent back reference resolution, on an example test
     scenario the runtime of send went down from a hour to minutes

   - factor out locking scheme used for subvolume writer and NOCOW
     exclusion, abstracted as DREW lock, double reader-writer exclusion
     (allow either readers or writers)

   - cleanup and abstract extent allocation policies, preparation for
     zoned device support

   - make reflink/clone_range work on inline extents

   - add more cancellation point for relocation, improves long response
     from 'balance cancel'

   - add page migration callback for data pages

   - switch to guid for uuids, with additional cleanups of the interface

   - make ranged full fsyncs more efficient

   - removal of obsolete ioctl flag BTRFS_SUBVOL_CREATE_ASYNC

   - remove b-tree readahead from delayed refs paths, avoiding seek and
     read unnecessary blocks

  Features:

   - v2 of ioctl to delete subvolumes, allowing to delete by id and more
     future extensions

  Fixes:

   - fix qgroup rescan worker that could block umount

   - fix crash during unmount due to race with delayed inode workers

   - fix dellaloc flushing logic that could create unnecessary chunks
     under heavy load

   - fix missing file extent item for hole after ranged fsync

   - several fixes in relocation error handling

  Other:

   - more documentation of relocation, device replace, space
     reservations

   - many random cleanups"

* tag 'for-5.7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (210 commits)
  btrfs: fix missing semaphore unlock in btrfs_sync_file
  btrfs: use nofs allocations for running delayed items
  btrfs: sysfs: Use scnprintf() instead of snprintf()
  btrfs: do not resolve backrefs for roots that are being deleted
  btrfs: track reloc roots based on their commit root bytenr
  btrfs: restart relocate_tree_blocks properly
  btrfs: reloc: reorder reservation before root selection
  btrfs: do not readahead in build_backref_tree
  btrfs: do not use readahead for running delayed refs
  btrfs: Remove async_transid from btrfs_mksubvol/create_subvol/create_snapshot
  btrfs: Remove transid argument from btrfs_ioctl_snap_create_transid
  btrfs: Remove BTRFS_SUBVOL_CREATE_ASYNC support
  btrfs: kill the subvol_srcu
  btrfs: make btrfs_cleanup_fs_roots use the radix tree lock
  btrfs: don't take an extra root ref at allocation time
  btrfs: hold a ref on the root on the dead roots list
  btrfs: make inodes hold a ref on their roots
  btrfs: move the root freeing stuff into btrfs_put_root
  btrfs: move ino_cache_inode dropping out of btrfs_free_fs_root
  btrfs: make the extent buffer leak check per fs info
  ...

fs/btrfs/Makefile

@@ -11,7 +11,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
 	   compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \
 	   reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o \
 	   uuid-tree.o props.o free-space-tree.o tree-checker.o space-info.o \
-	   block-rsv.o delalloc-space.o block-group.o discard.o
+	   block-rsv.o delalloc-space.o block-group.o discard.o reflink.o
 
 btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o
 btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o

fs/btrfs/async-thread.c

@@ -395,3 +395,11 @@ void btrfs_set_work_high_priority(struct btrfs_work *work)
 {
 	set_bit(WORK_HIGH_PRIO_BIT, &work->flags);
 }
+
+void btrfs_flush_workqueue(struct btrfs_workqueue *wq)
+{
+	if (wq->high)
+		flush_workqueue(wq->high->normal_wq);
+
+	flush_workqueue(wq->normal->normal_wq);
+}

fs/btrfs/async-thread.h

@@ -44,5 +44,6 @@ void btrfs_set_work_high_priority(struct btrfs_work *work);
 struct btrfs_fs_info * __pure btrfs_work_owner(const struct btrfs_work *work);
 struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct __btrfs_workqueue *wq);
 bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq);
+void btrfs_flush_workqueue(struct btrfs_workqueue *wq);
 
 #endif

fs/btrfs/backref.h

@@ -40,6 +40,10 @@ int iterate_inodes_from_logical(u64 logical, struct btrfs_fs_info *fs_info,
 
 int paths_from_inode(u64 inum, struct inode_fs_paths *ipath);
 
+int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
+			 struct btrfs_fs_info *fs_info, u64 bytenr,
+			 u64 time_seq, struct ulist **leafs,
+			 const u64 *extent_item_pos, bool ignore_offset);
 int btrfs_find_all_roots(struct btrfs_trans_handle *trans,
 			 struct btrfs_fs_info *fs_info, u64 bytenr,
 			 u64 time_seq, struct ulist **roots, bool ignore_offset);

fs/btrfs/block-group.c

@@ -460,7 +460,7 @@ u64 add_new_free_space(struct btrfs_block_group *block_group, u64 start, u64 end
 	int ret;
 
 	while (start < end) {
-		ret = find_first_extent_bit(info->pinned_extents, start,
+		ret = find_first_extent_bit(&info->excluded_extents, start,
 					    &extent_start, &extent_end,
 					    EXTENT_DIRTY | EXTENT_UPTODATE,
 					    NULL);
@@ -1248,6 +1248,55 @@ static int inc_block_group_ro(struct btrfs_block_group *cache, int force)
 	return ret;
 }
 
+static bool clean_pinned_extents(struct btrfs_trans_handle *trans,
+				 struct btrfs_block_group *bg)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+	struct btrfs_transaction *prev_trans = NULL;
+	const u64 start = bg->start;
+	const u64 end = start + bg->length - 1;
+	int ret;
+
+	spin_lock(&fs_info->trans_lock);
+	if (trans->transaction->list.prev != &fs_info->trans_list) {
+		prev_trans = list_last_entry(&trans->transaction->list,
+					     struct btrfs_transaction, list);
+		refcount_inc(&prev_trans->use_count);
+	}
+	spin_unlock(&fs_info->trans_lock);
+
+	/*
+	 * Hold the unused_bg_unpin_mutex lock to avoid racing with
+	 * btrfs_finish_extent_commit(). If we are at transaction N, another
+	 * task might be running finish_extent_commit() for the previous
+	 * transaction N - 1, and have seen a range belonging to the block
+	 * group in pinned_extents before we were able to clear the whole block
+	 * group range from pinned_extents. This means that task can lookup for
+	 * the block group after we unpinned it from pinned_extents and removed
+	 * it, leading to a BUG_ON() at unpin_extent_range().
+	 */
+	mutex_lock(&fs_info->unused_bg_unpin_mutex);
+	if (prev_trans) {
+		ret = clear_extent_bits(&prev_trans->pinned_extents, start, end,
+					EXTENT_DIRTY);
+		if (ret)
+			goto err;
+	}
+
+	ret = clear_extent_bits(&trans->transaction->pinned_extents, start, end,
+				EXTENT_DIRTY);
+	if (ret)
+		goto err;
+	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+
+	return true;
+
+err:
+	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+	btrfs_dec_block_group_ro(bg);
+	return false;
+}
+
 /*
  * Process the unused_bgs list and remove any that don't have any allocated
  * space inside of them.
@@ -1265,7 +1314,6 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 	spin_lock(&fs_info->unused_bgs_lock);
 	while (!list_empty(&fs_info->unused_bgs)) {
-		u64 start, end;
 		int trimming;
 
 		block_group = list_first_entry(&fs_info->unused_bgs,
@@ -1344,35 +1392,8 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * We could have pending pinned extents for this block group,
 		 * just delete them, we don't care about them anymore.
 		 */
-		start = block_group->start;
-		end = start + block_group->length - 1;
-		/*
-		 * Hold the unused_bg_unpin_mutex lock to avoid racing with
-		 * btrfs_finish_extent_commit(). If we are at transaction N,
-		 * another task might be running finish_extent_commit() for the
-		 * previous transaction N - 1, and have seen a range belonging
-		 * to the block group in freed_extents[] before we were able to
-		 * clear the whole block group range from freed_extents[]. This
-		 * means that task can lookup for the block group after we
-		 * unpinned it from freed_extents[] and removed it, leading to
-		 * a BUG_ON() at btrfs_unpin_extent_range().
-		 */
-		mutex_lock(&fs_info->unused_bg_unpin_mutex);
-		ret = clear_extent_bits(&fs_info->freed_extents[0], start, end,
-				  EXTENT_DIRTY);
-		if (ret) {
-			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-			btrfs_dec_block_group_ro(block_group);
+		if (!clean_pinned_extents(trans, block_group))
 			goto end_trans;
-		}
-		ret = clear_extent_bits(&fs_info->freed_extents[1], start, end,
-				  EXTENT_DIRTY);
-		if (ret) {
-			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-			btrfs_dec_block_group_ro(block_group);
-			goto end_trans;
-		}
-		mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 
 		/*
 		 * At this point, the block_group is read only and should fail
@@ -1987,6 +2008,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		btrfs_release_path(path);
 	}
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(space_info, &info->space_info, list) {
 		if (!(btrfs_get_alloc_profile(info, space_info->flags) &
 		      (BTRFS_BLOCK_GROUP_RAID10 |
@@ -2007,6 +2029,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 				list)
 			inc_block_group_ro(cache, 1);
 	}
+	rcu_read_unlock();
 
 	btrfs_init_global_block_rsv(info);
 	ret = check_chunk_block_group_mappings(info);
@@ -2345,7 +2368,7 @@ static int cache_save_setup(struct btrfs_block_group *block_group,
 		return 0;
 	}
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 again:
 	inode = lookup_free_space_inode(block_group, path);
@@ -2881,7 +2904,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 					&cache->space_info->total_bytes_pinned,
 					num_bytes,
 					BTRFS_TOTAL_BYTES_PINNED_BATCH);
-			set_extent_dirty(info->pinned_extents,
+			set_extent_dirty(&trans->transaction->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
 		}

fs/btrfs/block-rsv.c

@@ -6,6 +6,98 @@
 #include "space-info.h"
 #include "transaction.h"
 
+/*
+ * HOW DO BLOCK RESERVES WORK
+ *
+ *   Think of block_rsv's as buckets for logically grouped metadata
+ *   reservations.  Each block_rsv has a ->size and a ->reserved.  ->size is
+ *   how large we want our block rsv to be, ->reserved is how much space is
+ *   currently reserved for this block reserve.
+ *
+ *   ->failfast exists for the truncate case, and is described below.
+ *
+ * NORMAL OPERATION
+ *
+ *   -> Reserve
+ *     Entrance: btrfs_block_rsv_add, btrfs_block_rsv_refill
+ *
+ *     We call into btrfs_reserve_metadata_bytes() with our bytes, which is
+ *     accounted for in space_info->bytes_may_use, and then add the bytes to
+ *     ->reserved, and ->size in the case of btrfs_block_rsv_add.
+ *
+ *     ->size is an over-estimation of how much we may use for a particular
+ *     operation.
+ *
+ *   -> Use
+ *     Entrance: btrfs_use_block_rsv
+ *
+ *     When we do a btrfs_alloc_tree_block() we call into btrfs_use_block_rsv()
+ *     to determine the appropriate block_rsv to use, and then verify that
+ *     ->reserved has enough space for our tree block allocation.  Once
+ *     successful we subtract fs_info->nodesize from ->reserved.
+ *
+ *   -> Finish
+ *     Entrance: btrfs_block_rsv_release
+ *
+ *     We are finished with our operation, subtract our individual reservation
+ *     from ->size, and then subtract ->size from ->reserved and free up the
+ *     excess if there is any.
+ *
+ *     There is some logic here to refill the delayed refs rsv or the global rsv
+ *     as needed, otherwise the excess is subtracted from
+ *     space_info->bytes_may_use.
+ *
+ * TYPES OF BLOCK RESERVES
+ *
+ * BLOCK_RSV_TRANS, BLOCK_RSV_DELOPS, BLOCK_RSV_CHUNK
+ *   These behave normally, as described above, just within the confines of the
+ *   lifetime of their particular operation (transaction for the whole trans
+ *   handle lifetime, for example).
+ *
+ * BLOCK_RSV_GLOBAL
+ *   It is impossible to properly account for all the space that may be required
+ *   to make our extent tree updates.  This block reserve acts as an overflow
+ *   buffer in case our delayed refs reserve does not reserve enough space to
+ *   update the extent tree.
+ *
+ *   We can steal from this in some cases as well, notably on evict() or
+ *   truncate() in order to help users recover from ENOSPC conditions.
+ *
+ * BLOCK_RSV_DELALLOC
+ *   The individual item sizes are determined by the per-inode size
+ *   calculations, which are described with the delalloc code.  This is pretty
+ *   straightforward, it's just the calculation of ->size encodes a lot of
+ *   different items, and thus it gets used when updating inodes, inserting file
+ *   extents, and inserting checksums.
+ *
+ * BLOCK_RSV_DELREFS
+ *   We keep a running tally of how many delayed refs we have on the system.
+ *   We assume each one of these delayed refs are going to use a full
+ *   reservation.  We use the transaction items and pre-reserve space for every
+ *   operation, and use this reservation to refill any gap between ->size and
+ *   ->reserved that may exist.
+ *
+ *   From there it's straightforward, removing a delayed ref means we remove its
+ *   count from ->size and free up reservations as necessary.  Since this is
+ *   the most dynamic block reserve in the system, we will try to refill this
+ *   block reserve first with any excess returned by any other block reserve.
+ *
+ * BLOCK_RSV_EMPTY
+ *   This is the fallback block reserve to make us try to reserve space if we
+ *   don't have a specific bucket for this allocation.  It is mostly used for
+ *   updating the device tree and such, since that is a separate pool we're
+ *   content to just reserve space from the space_info on demand.
+ *
+ * BLOCK_RSV_TEMP
+ *   This is used by things like truncate and iput.  We will temporarily
+ *   allocate a block reserve, set it to some size, and then truncate bytes
+ *   until we have no space left.  With ->failfast set we'll simply return
+ *   ENOSPC from btrfs_use_block_rsv() to signal that we need to unwind and try
+ *   to make a new reservation.  This is because these operations are
+ *   unbounded, so we want to do as much work as we can, and then back off and
+ *   re-reserve.
+ */
+
 static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_rsv *block_rsv,
 				    struct btrfs_block_rsv *dest, u64 num_bytes,
@@ -111,7 +203,7 @@ void btrfs_free_block_rsv(struct btrfs_fs_info *fs_info,
 {
 	if (!rsv)
 		return;
-	btrfs_block_rsv_release(fs_info, rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, rsv, (u64)-1, NULL);
 	kfree(rsv);
 }
 
@@ -178,9 +270,9 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 	return ret;
 }
 
-u64 __btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
-			      struct btrfs_block_rsv *block_rsv,
-			      u64 num_bytes, u64 *qgroup_to_release)
+u64 btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
+			    struct btrfs_block_rsv *block_rsv, u64 num_bytes,
+			    u64 *qgroup_to_release)
 {
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
 	struct btrfs_block_rsv *delayed_rsv = &fs_info->delayed_refs_rsv;
@@ -297,9 +389,9 @@ void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info)
 
 	if (block_rsv->reserved < block_rsv->size) {
 		num_bytes = block_rsv->size - block_rsv->reserved;
-		block_rsv->reserved += num_bytes;
 		btrfs_space_info_update_bytes_may_use(fs_info, sinfo,
 						      num_bytes);
+		block_rsv->reserved = block_rsv->size;
 	} else if (block_rsv->reserved > block_rsv->size) {
 		num_bytes = block_rsv->reserved - block_rsv->size;
 		btrfs_space_info_update_bytes_may_use(fs_info, sinfo,
@@ -344,7 +436,8 @@ void btrfs_init_global_block_rsv(struct btrfs_fs_info *fs_info)
 
 void btrfs_release_global_block_rsv(struct btrfs_fs_info *fs_info)
 {
-	btrfs_block_rsv_release(fs_info, &fs_info->global_block_rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, &fs_info->global_block_rsv, (u64)-1,
+				NULL);
 	WARN_ON(fs_info->trans_block_rsv.size > 0);
 	WARN_ON(fs_info->trans_block_rsv.reserved > 0);
 	WARN_ON(fs_info->chunk_block_rsv.size > 0);

fs/btrfs/block-rsv.h

@@ -73,7 +73,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 			     int min_factor);
 void btrfs_block_rsv_add_bytes(struct btrfs_block_rsv *block_rsv,
 			       u64 num_bytes, bool update_size);
-u64 __btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
+u64 btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
 			      struct btrfs_block_rsv *block_rsv,
 			      u64 num_bytes, u64 *qgroup_to_release);
 void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info);
@@ -82,20 +82,12 @@ void btrfs_release_global_block_rsv(struct btrfs_fs_info *fs_info);
 struct btrfs_block_rsv *btrfs_use_block_rsv(struct btrfs_trans_handle *trans,
 					    struct btrfs_root *root,
 					    u32 blocksize);
-
-static inline void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
-					   struct btrfs_block_rsv *block_rsv,
-					   u64 num_bytes)
-{
-	__btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, NULL);
-}
-
 static inline void btrfs_unuse_block_rsv(struct btrfs_fs_info *fs_info,
 					 struct btrfs_block_rsv *block_rsv,
 					 u32 blocksize)
 {
 	btrfs_block_rsv_add_bytes(block_rsv, blocksize, false);
-	btrfs_block_rsv_release(fs_info, block_rsv, 0);
+	btrfs_block_rsv_release(fs_info, block_rsv, 0, NULL);
 }
 
 #endif /* BTRFS_BLOCK_RSV_H */

fs/btrfs/btrfs_inode.h

@@ -60,6 +60,12 @@ struct btrfs_inode {
 	 */
 	struct extent_io_tree io_failure_tree;
 
+	/*
+	 * Keep track of where the inode has extent items mapped in order to
+	 * make sure the i_size adjustments are accurate
+	 */
+	struct extent_io_tree file_extent_tree;
+
 	/* held while logging the inode in tree-log.c */
 	struct mutex log_mutex;
 

fs/btrfs/check-integrity.h

@@ -7,11 +7,9 @@
 #define BTRFS_CHECK_INTEGRITY_H
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh);
 void btrfsic_submit_bio(struct bio *bio);
 int btrfsic_submit_bio_wait(struct bio *bio);
 #else
-#define btrfsic_submit_bh submit_bh
 #define btrfsic_submit_bio submit_bio
 #define btrfsic_submit_bio_wait submit_bio_wait
 #endif

fs/btrfs/ctree.c

@@ -31,8 +31,8 @@ static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
 
 static const struct btrfs_csums {
 	u16		size;
-	const char	*name;
-	const char	*driver;
+	const char	name[10];
+	const char	driver[12];
 } btrfs_csums[] = {
 	[BTRFS_CSUM_TYPE_CRC32] = { .size = 4, .name = "crc32c" },
 	[BTRFS_CSUM_TYPE_XXHASH] = { .size = 8, .name = "xxhash64" },
@@ -63,7 +63,8 @@ const char *btrfs_super_csum_name(u16 csum_type)
 const char *btrfs_super_csum_driver(u16 csum_type)
 {
 	/* csum type is validated at mount time */
-	return btrfs_csums[csum_type].driver ?:
+	return btrfs_csums[csum_type].driver[0] ?
+		btrfs_csums[csum_type].driver :
 		btrfs_csums[csum_type].name;
 }
 
@@ -143,44 +144,6 @@ struct extent_buffer *btrfs_root_node(struct btrfs_root *root)
 	return eb;
 }
 
-/* loop around taking references on and locking the root node of the
- * tree until you end up with a lock on the root.  A locked buffer
- * is returned, with a reference held.
- */
-struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
-{
-	struct extent_buffer *eb;
-
-	while (1) {
-		eb = btrfs_root_node(root);
-		btrfs_tree_lock(eb);
-		if (eb == root->node)
-			break;
-		btrfs_tree_unlock(eb);
-		free_extent_buffer(eb);
-	}
-	return eb;
-}
-
-/* loop around taking references on and locking the root node of the
- * tree until you end up with a lock on the root.  A locked buffer
- * is returned, with a reference held.
- */
-struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
-{
-	struct extent_buffer *eb;
-
-	while (1) {
-		eb = btrfs_root_node(root);
-		btrfs_tree_read_lock(eb);
-		if (eb == root->node)
-			break;
-		btrfs_tree_read_unlock(eb);
-		free_extent_buffer(eb);
-	}
-	return eb;
-}
-
 /* cowonly root (everything not a reference counted cow subvolume), just get
  * put onto a simple dirty list.  transaction.c walks this to make sure they
  * get properly updated on disk.
@@ -341,7 +304,6 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 	struct rb_root *tm_root;
 	struct rb_node *node;
 	struct rb_node *next;
-	struct seq_list *cur_elem;
 	struct tree_mod_elem *tm;
 	u64 min_seq = (u64)-1;
 	u64 seq_putting = elem->seq;
@@ -353,18 +315,20 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 	list_del(&elem->list);
 	elem->seq = 0;
 
-	list_for_each_entry(cur_elem, &fs_info->tree_mod_seq_list, list) {
-		if (cur_elem->seq < min_seq) {
-			if (seq_putting > cur_elem->seq) {
-				/*
-				 * blocker with lower sequence number exists, we
-				 * cannot remove anything from the log
-				 */
-				write_unlock(&fs_info->tree_mod_log_lock);
-				return;
-			}
-			min_seq = cur_elem->seq;
+	if (!list_empty(&fs_info->tree_mod_seq_list)) {
+		struct seq_list *first;
+
+		first = list_first_entry(&fs_info->tree_mod_seq_list,
+					 struct seq_list, list);
+		if (seq_putting > first->seq) {
+			/*
+			 * Blocker with lower sequence number exists, we
+			 * cannot remove anything from the log.
+			 */
+			write_unlock(&fs_info->tree_mod_log_lock);
+			return;
 		}
+		min_seq = first->seq;
 	}
 
 	/*
@@ -962,9 +926,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 		if (new_flags != 0) {
 			int level = btrfs_header_level(buf);
 
-			ret = btrfs_set_disk_extent_flags(trans,
-							  buf->start,
-							  buf->len,
+			ret = btrfs_set_disk_extent_flags(trans, buf,
 							  new_flags, level, 0);
 			if (ret)
 				return ret;

fs/btrfs/ctree.h

@@ -33,6 +33,7 @@
 #include "extent_map.h"
 #include "async-thread.h"
 #include "block-rsv.h"
+#include "locking.h"
 
 struct btrfs_trans_handle;
 struct btrfs_transaction;
@@ -596,8 +597,8 @@ struct btrfs_fs_info {
 	/* keep track of unallocated space */
 	atomic64_t free_chunk_space;
 
-	struct extent_io_tree freed_extents[2];
-	struct extent_io_tree *pinned_extents;
+	/* Track ranges which are used by log trees blocks/logged data extents */
+	struct extent_io_tree excluded_extents;
 
 	/* logical->physical extent mapping */
 	struct extent_map_tree mapping_tree;
@@ -696,7 +697,6 @@ struct btrfs_fs_info {
 	struct rw_semaphore cleanup_work_sem;
 
 	struct rw_semaphore subvol_sem;
-	struct srcu_struct subvol_srcu;
 
 	spinlock_t trans_lock;
 	/*
@@ -947,6 +947,10 @@ struct btrfs_fs_info {
 #ifdef CONFIG_BTRFS_DEBUG
 	struct kobject *debug_kobj;
 	struct kobject *discard_debug_kobj;
+	struct list_head allocated_roots;
+
+	spinlock_t eb_leak_lock;
+	struct list_head allocated_ebs;
 #endif
 };
 
@@ -955,11 +959,6 @@ static inline struct btrfs_fs_info *btrfs_sb(struct super_block *sb)
 	return sb->s_fs_info;
 }
 
-struct btrfs_subvolume_writers {
-	struct percpu_counter	counter;
-	wait_queue_head_t	wait;
-};
-
 /*
  * The state of btrfs root
  */
@@ -1131,8 +1130,9 @@ struct btrfs_root {
 	 * root_item_lock.
 	 */
 	int dedupe_in_progress;
-	struct btrfs_subvolume_writers *subv_writers;
-	atomic_t will_be_snapshotted;
+	/* For exclusion of snapshot creation and nocow writes */
+	struct btrfs_drew_lock snapshot_lock;
+
 	atomic_t snapshot_force_cow;
 
 	/* For qgroup metadata reserved space */
@@ -1149,6 +1149,10 @@ struct btrfs_root {
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	u64 alloc_bytenr;
 #endif
+
+#ifdef CONFIG_BTRFS_DEBUG
+	struct list_head leak_list;
+#endif
 };
 
 struct btrfs_clone_extent_info {
@@ -1971,16 +1975,6 @@ static inline void btrfs_set_header_backref_rev(struct extent_buffer *eb,
 	btrfs_set_header_flags(eb, flags);
 }
 
-static inline unsigned long btrfs_header_fsid(void)
-{
-	return offsetof(struct btrfs_header, fsid);
-}
-
-static inline unsigned long btrfs_header_chunk_tree_uuid(const struct extent_buffer *eb)
-{
-	return offsetof(struct btrfs_header, chunk_tree_uuid);
-}
-
 static inline int btrfs_is_leaf(const struct extent_buffer *eb)
 {
 	return btrfs_header_level(eb) == 0;
@@ -2458,9 +2452,9 @@ int btrfs_lookup_data_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len);
 int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
 			     struct btrfs_fs_info *fs_info, u64 bytenr,
 			     u64 offset, int metadata, u64 *refs, u64 *flags);
-int btrfs_pin_extent(struct btrfs_fs_info *fs_info,
-		     u64 bytenr, u64 num, int reserved);
-int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info,
+int btrfs_pin_extent(struct btrfs_trans_handle *trans, u64 bytenr, u64 num,
+		     int reserved);
+int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
 				    u64 bytenr, u64 num_bytes);
 int btrfs_exclude_logged_extents(struct extent_buffer *eb);
 int btrfs_cross_ref_exist(struct btrfs_root *root,
@@ -2490,13 +2484,13 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		  struct extent_buffer *buf, int full_backref);
 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
-				u64 bytenr, u64 num_bytes, u64 flags,
+				struct extent_buffer *eb, u64 flags,
 				int level, int is_data);
 int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref);
 
 int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
 			       u64 start, u64 len, int delalloc);
-int btrfs_pin_reserved_extent(struct btrfs_fs_info *fs_info, u64 start,
+int btrfs_pin_reserved_extent(struct btrfs_trans_handle *trans, u64 start,
 			      u64 len);
 void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info);
 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans);
@@ -2665,9 +2659,8 @@ static inline int btrfs_next_item(struct btrfs_root *root, struct btrfs_path *p)
 	return btrfs_next_old_item(root, p, 0);
 }
 int btrfs_leaf_free_space(struct extent_buffer *leaf);
-int __must_check btrfs_drop_snapshot(struct btrfs_root *root,
-				     struct btrfs_block_rsv *block_rsv,
-				     int update_ref, int for_reloc);
+int __must_check btrfs_drop_snapshot(struct btrfs_root *root, int update_ref,
+				     int for_reloc);
 int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root,
 			struct extent_buffer *node,
@@ -2695,23 +2688,6 @@ static inline int btrfs_need_cleaner_sleep(struct btrfs_fs_info *fs_info)
 	return fs_info->sb->s_flags & SB_RDONLY || btrfs_fs_closing(fs_info);
 }
 
-static inline void free_fs_info(struct btrfs_fs_info *fs_info)
-{
-	kfree(fs_info->balance_ctl);
-	kfree(fs_info->delayed_root);
-	kfree(fs_info->extent_root);
-	kfree(fs_info->tree_root);
-	kfree(fs_info->chunk_root);
-	kfree(fs_info->dev_root);
-	kfree(fs_info->csum_root);
-	kfree(fs_info->quota_root);
-	kfree(fs_info->uuid_root);
-	kfree(fs_info->free_space_root);
-	kfree(fs_info->super_copy);
-	kfree(fs_info->super_for_commit);
-	kvfree(fs_info);
-}
-
 /* tree mod log functions from ctree.c */
 u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info,
 			   struct seq_list *elem);
@@ -2750,9 +2726,7 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid);
 int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid);
-int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
-			    int (*check_func)(struct btrfs_fs_info *, u8 *, u8,
-					      u64));
+int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info);
 
 /* dir-item.c */
 int btrfs_check_dir_item_collision(struct btrfs_root *root, u64 dir,
@@ -2859,6 +2833,12 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 				     struct btrfs_file_extent_item *fi,
 				     const bool new_inline,
 				     struct extent_map *em);
+int btrfs_inode_clear_file_extent_range(struct btrfs_inode *inode, u64 start,
+					u64 len);
+int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start,
+				      u64 len);
+void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size);
+u64 btrfs_file_extent_end(const struct btrfs_path *path);
 
 /* inode.c */
 struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
@@ -2996,9 +2976,6 @@ int btrfs_dirty_pages(struct inode *inode, struct page **pages,
 		      size_t num_pages, loff_t pos, size_t write_bytes,
 		      struct extent_state **cached);
 int btrfs_fdatawrite_range(struct inode *inode, loff_t start, loff_t end);
-loff_t btrfs_remap_file_range(struct file *file_in, loff_t pos_in,
-			      struct file *file_out, loff_t pos_out,
-			      loff_t len, unsigned int remap_flags);
 
 /* tree-defrag.c */
 int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
@@ -3008,6 +2985,8 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 			unsigned long new_flags);
 int btrfs_sync_fs(struct super_block *sb, int wait);
+char *btrfs_get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
+					  u64 subvol_objectid);
 
 static inline __printf(2, 3) __cold
 void btrfs_no_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
@@ -3401,6 +3380,7 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
 			      u64 *bytes_to_reserve);
 int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 			      struct btrfs_pending_snapshot *pending);
+int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info);
 
 /* scrub.c */
 int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,

fs/btrfs/delalloc-space.c

@@ -9,6 +9,108 @@
 #include "qgroup.h"
 #include "block-group.h"
 
+/*
+ * HOW DOES THIS WORK
+ *
+ * There are two stages to data reservations, one for data and one for metadata
+ * to handle the new extents and checksums generated by writing data.
+ *
+ *
+ * DATA RESERVATION
+ *   The general flow of the data reservation is as follows
+ *
+ *   -> Reserve
+ *     We call into btrfs_reserve_data_bytes() for the user request bytes that
+ *     they wish to write.  We make this reservation and add it to
+ *     space_info->bytes_may_use.  We set EXTENT_DELALLOC on the inode io_tree
+ *     for the range and carry on if this is buffered, or follow up trying to
+ *     make a real allocation if we are pre-allocating or doing O_DIRECT.
+ *
+ *   -> Use
+ *     At writepages()/prealloc/O_DIRECT time we will call into
+ *     btrfs_reserve_extent() for some part or all of this range of bytes.  We
+ *     will make the allocation and subtract space_info->bytes_may_use by the
+ *     original requested length and increase the space_info->bytes_reserved by
+ *     the allocated length.  This distinction is important because compression
+ *     may allocate a smaller on disk extent than we previously reserved.
+ *
+ *   -> Allocation
+ *     finish_ordered_io() will insert the new file extent item for this range,
+ *     and then add a delayed ref update for the extent tree.  Once that delayed
+ *     ref is written the extent size is subtracted from
+ *     space_info->bytes_reserved and added to space_info->bytes_used.
+ *
+ *   Error handling
+ *
+ *   -> By the reservation maker
+ *     This is the simplest case, we haven't completed our operation and we know
+ *     how much we reserved, we can simply call
+ *     btrfs_free_reserved_data_space*() and it will be removed from
+ *     space_info->bytes_may_use.
+ *
+ *   -> After the reservation has been made, but before cow_file_range()
+ *     This is specifically for the delalloc case.  You must clear
+ *     EXTENT_DELALLOC with the EXTENT_CLEAR_DATA_RESV bit, and the range will
+ *     be subtracted from space_info->bytes_may_use.
+ *
+ * METADATA RESERVATION
+ *   The general metadata reservation lifetimes are discussed elsewhere, this
+ *   will just focus on how it is used for delalloc space.
+ *
+ *   We keep track of two things on a per inode bases
+ *
+ *   ->outstanding_extents
+ *     This is the number of file extent items we'll need to handle all of the
+ *     outstanding DELALLOC space we have in this inode.  We limit the maximum
+ *     size of an extent, so a large contiguous dirty area may require more than
+ *     one outstanding_extent, which is why count_max_extents() is used to
+ *     determine how many outstanding_extents get added.
+ *
+ *   ->csum_bytes
+ *     This is essentially how many dirty bytes we have for this inode, so we
+ *     can calculate the number of checksum items we would have to add in order
+ *     to checksum our outstanding data.
+ *
+ *   We keep a per-inode block_rsv in order to make it easier to keep track of
+ *   our reservation.  We use btrfs_calculate_inode_block_rsv_size() to
+ *   calculate the current theoretical maximum reservation we would need for the
+ *   metadata for this inode.  We call this and then adjust our reservation as
+ *   necessary, either by attempting to reserve more space, or freeing up excess
+ *   space.
+ *
+ * OUTSTANDING_EXTENTS HANDLING
+ *
+ *  ->outstanding_extents is used for keeping track of how many extents we will
+ *  need to use for this inode, and it will fluctuate depending on where you are
+ *  in the life cycle of the dirty data.  Consider the following normal case for
+ *  a completely clean inode, with a num_bytes < our maximum allowed extent size
+ *
+ *  -> reserve
+ *    ->outstanding_extents += 1 (current value is 1)
+ *
+ *  -> set_delalloc
+ *    ->outstanding_extents += 1 (currrent value is 2)
+ *
+ *  -> btrfs_delalloc_release_extents()
+ *    ->outstanding_extents -= 1 (current value is 1)
+ *
+ *    We must call this once we are done, as we hold our reservation for the
+ *    duration of our operation, and then assume set_delalloc will update the
+ *    counter appropriately.
+ *
+ *  -> add ordered extent
+ *    ->outstanding_extents += 1 (current value is 2)
+ *
+ *  -> btrfs_clear_delalloc_extent
+ *    ->outstanding_extents -= 1 (current value is 1)
+ *
+ *  -> finish_ordered_io/btrfs_remove_ordered_extent
+ *    ->outstanding_extents -= 1 (current value is 0)
+ *
+ *  Each stage is responsible for their own accounting of the extent, thus
+ *  making error handling and cleanup easier.
+ */
+
 int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes)
 {
 	struct btrfs_root *root = inode->root;
@@ -228,8 +330,8 @@ static void btrfs_inode_rsv_release(struct btrfs_inode *inode, bool qgroup_free)
 	 * are releasing 0 bytes, and then we'll just get the reservation over
 	 * the size free'd.
 	 */
-	released = __btrfs_block_rsv_release(fs_info, block_rsv, 0,
-					     &qgroup_to_release);
+	released = btrfs_block_rsv_release(fs_info, block_rsv, 0,
+					   &qgroup_to_release);
 	if (released > 0)
 		trace_btrfs_space_reservation(fs_info, "delalloc",
 					      btrfs_ino(inode), released, 0);

fs/btrfs/delayed-inode.c

@@ -6,6 +6,7 @@
 
 #include <linux/slab.h>
 #include <linux/iversion.h>
+#include <linux/sched/mm.h>
 #include "misc.h"
 #include "delayed-inode.h"
 #include "disk-io.h"
@@ -595,8 +596,7 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
 	trace_btrfs_space_reservation(fs_info, "delayed_item",
 				      item->key.objectid, item->bytes_reserved,
 				      0);
-	btrfs_block_rsv_release(fs_info, rsv,
-				item->bytes_reserved);
+	btrfs_block_rsv_release(fs_info, rsv, item->bytes_reserved, NULL);
 }
 
 static int btrfs_delayed_inode_reserve_metadata(
@@ -677,8 +677,7 @@ static void btrfs_delayed_inode_release_metadata(struct btrfs_fs_info *fs_info,
 	rsv = &fs_info->delayed_block_rsv;
 	trace_btrfs_space_reservation(fs_info, "delayed_inode",
 				      node->inode_id, node->bytes_reserved, 0);
-	btrfs_block_rsv_release(fs_info, rsv,
-				node->bytes_reserved);
+	btrfs_block_rsv_release(fs_info, rsv, node->bytes_reserved, NULL);
 	if (qgroup_free)
 		btrfs_qgroup_free_meta_prealloc(node->root,
 				node->bytes_reserved);
@@ -805,11 +804,14 @@ static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans,
 				     struct btrfs_delayed_item *delayed_item)
 {
 	struct extent_buffer *leaf;
+	unsigned int nofs_flag;
 	char *ptr;
 	int ret;
 
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_insert_empty_item(trans, root, path, &delayed_item->key,
 				      delayed_item->data_len);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret < 0 && ret != -EEXIST)
 		return ret;
 
@@ -937,6 +939,7 @@ static int btrfs_delete_delayed_items(struct btrfs_trans_handle *trans,
 				      struct btrfs_delayed_node *node)
 {
 	struct btrfs_delayed_item *curr, *prev;
+	unsigned int nofs_flag;
 	int ret = 0;
 
 do_again:
@@ -945,7 +948,9 @@ static int btrfs_delete_delayed_items(struct btrfs_trans_handle *trans,
 	if (!curr)
 		goto delete_fail;
 
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_search_slot(trans, root, &curr->key, path, -1, 1);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret < 0)
 		goto delete_fail;
 	else if (ret > 0) {
@@ -1012,6 +1017,7 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
 	struct btrfs_key key;
 	struct btrfs_inode_item *inode_item;
 	struct extent_buffer *leaf;
+	unsigned int nofs_flag;
 	int mod;
 	int ret;
 
@@ -1024,7 +1030,9 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
 	else
 		mod = 1;
 
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_lookup_inode(trans, root, path, &key, mod);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret > 0) {
 		btrfs_release_path(path);
 		return -ENOENT;
@@ -1075,7 +1083,10 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
 
 	key.type = BTRFS_INODE_EXTREF_KEY;
 	key.offset = -1;
+
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret < 0)
 		goto err_out;
 	ASSERT(ret);
@@ -1139,7 +1150,7 @@ static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans, int nr)
 	int ret = 0;
 	bool count = (nr > 0);
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return -EIO;
 
 	path = btrfs_alloc_path();
@@ -1760,6 +1771,7 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
 
 int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 {
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
 	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_inode_item *inode_item;
 
@@ -1779,6 +1791,8 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 	i_uid_write(inode, btrfs_stack_inode_uid(inode_item));
 	i_gid_write(inode, btrfs_stack_inode_gid(inode_item));
 	btrfs_i_size_write(BTRFS_I(inode), btrfs_stack_inode_size(inode_item));
+	btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0,
+			round_up(i_size_read(inode), fs_info->sectorsize));
 	inode->i_mode = btrfs_stack_inode_mode(inode_item);
 	set_nlink(inode, btrfs_stack_inode_nlink(inode_item));
 	inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item));

fs/btrfs/delayed-inode.h

@@ -70,7 +70,7 @@ struct btrfs_delayed_item {
 	refcount_t refs;
 	int ins_or_del;
 	u32 data_len;
-	char data[0];
+	char data[];
 };
 
 static inline void btrfs_init_delayed_root(

fs/btrfs/delayed-ref.c

@@ -82,8 +82,7 @@ void btrfs_delayed_refs_rsv_release(struct btrfs_fs_info *fs_info, int nr)
 	u64 num_bytes = btrfs_calc_insert_metadata_size(fs_info, nr);
 	u64 released = 0;
 
-	released = __btrfs_block_rsv_release(fs_info, block_rsv, num_bytes,
-					     NULL);
+	released = btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, NULL);
 	if (released)
 		trace_btrfs_space_reservation(fs_info, "delayed_refs_rsv",
 					      0, released, 0);

fs/btrfs/dev-replace.c

@@ -22,6 +22,46 @@
 #include "dev-replace.h"
 #include "sysfs.h"
 
+/*
+ * Device replace overview
+ *
+ * [Objective]
+ * To copy all extents (both new and on-disk) from source device to target
+ * device, while still keeping the filesystem read-write.
+ *
+ * [Method]
+ * There are two main methods involved:
+ *
+ * - Write duplication
+ *
+ *   All new writes will be written to both target and source devices, so even
+ *   if replace gets canceled, sources device still contans up-to-date data.
+ *
+ *   Location:		handle_ops_on_dev_replace() from __btrfs_map_block()
+ *   Start:		btrfs_dev_replace_start()
+ *   End:		btrfs_dev_replace_finishing()
+ *   Content:		Latest data/metadata
+ *
+ * - Copy existing extents
+ *
+ *   This happens by re-using scrub facility, as scrub also iterates through
+ *   existing extents from commit root.
+ *
+ *   Location:		scrub_write_block_to_dev_replace() from
+ *   			scrub_block_complete()
+ *   Content:		Data/meta from commit root.
+ *
+ * Due to the content difference, we need to avoid nocow write when dev-replace
+ * is happening.  This is done by marking the block group read-only and waiting
+ * for NOCOW writes.
+ *
+ * After replace is done, the finishing part is done by swapping the target and
+ * source devices.
+ *
+ *   Location:		btrfs_dev_replace_update_device_in_mapping_tree() from
+ *   			btrfs_dev_replace_finishing()
+ */
+
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 				       int scrub_ret);
 static void btrfs_dev_replace_update_device_in_mapping_tree(
@@ -472,7 +512,7 @@ static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
 	up_write(&dev_replace->rwsem);
 
-	ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
+	ret = btrfs_sysfs_add_devices_dir(tgt_device->fs_devices, tgt_device);
 	if (ret)
 		btrfs_err(fs_info, "kobj add dev failed %d", ret);
 
@@ -703,7 +743,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	/* replace the sysfs entry */
-	btrfs_sysfs_rm_device_link(fs_info->fs_devices, src_device);
+	btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, src_device);
 	btrfs_sysfs_update_devid(tgt_device);
 	btrfs_rm_dev_replace_free_srcdev(src_device);
 

fs/btrfs/disk-io.h

@@ -39,6 +39,8 @@ static inline u64 btrfs_sb_offset(int mirror)
 struct btrfs_device;
 struct btrfs_fs_devices;
 
+void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info);
+void btrfs_init_fs_info(struct btrfs_fs_info *fs_info);
 int btrfs_verify_level_key(struct extent_buffer *eb, int level,
 			   struct btrfs_key *first_key, u64 parent_transid);
 struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
@@ -54,15 +56,12 @@ int __cold open_ctree(struct super_block *sb,
 	       char *options);
 void __cold close_ctree(struct btrfs_fs_info *fs_info);
 int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors);
-struct buffer_head *btrfs_read_dev_super(struct block_device *bdev);
-int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num,
-			struct buffer_head **bh_ret);
+struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev);
+struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
+						   int copy_num);
 int btrfs_commit_super(struct btrfs_fs_info *fs_info);
-struct btrfs_root *btrfs_read_fs_root(struct btrfs_root *tree_root,
-				      struct btrfs_key *location);
-int btrfs_init_fs_root(struct btrfs_root *root);
-struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
-					u64 root_id);
+struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
+					struct btrfs_key *key);
 int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 			 struct btrfs_root *root);
 void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info);
@@ -70,19 +69,13 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info);
 struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
 				     struct btrfs_key *key,
 				     bool check_ref);
-static inline struct btrfs_root *
-btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info,
-			   struct btrfs_key *location)
-{
-	return btrfs_get_fs_root(fs_info, location, true);
-}
 
+void btrfs_free_fs_info(struct btrfs_fs_info *fs_info);
 int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info);
 void btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info);
 void btrfs_btree_balance_dirty_nodelay(struct btrfs_fs_info *fs_info);
 void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 				 struct btrfs_root *root);
-void btrfs_free_fs_root(struct btrfs_root *root);
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
@@ -95,19 +88,16 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
  * If you want to ensure the whole tree is safe, you should use
  * 	fs_info->subvol_srcu
  */
-static inline struct btrfs_root *btrfs_grab_fs_root(struct btrfs_root *root)
+static inline struct btrfs_root *btrfs_grab_root(struct btrfs_root *root)
 {
+	if (!root)
+		return NULL;
 	if (refcount_inc_not_zero(&root->refs))
 		return root;
 	return NULL;
 }
 
-static inline void btrfs_put_fs_root(struct btrfs_root *root)
-{
-	if (refcount_dec_and_test(&root->refs))
-		kfree(root);
-}
-
+void btrfs_put_root(struct btrfs_root *root);
 void btrfs_mark_buffer_dirty(struct extent_buffer *buf);
 int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
 			  int atomic);

fs/btrfs/export.c

@@ -57,16 +57,14 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
 	return type;
 }
 
-static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
-				       u64 root_objectid, u32 generation,
-				       int check_generation)
+struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
+				u64 root_objectid, u32 generation,
+				int check_generation)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
 	struct btrfs_root *root;
 	struct inode *inode;
 	struct btrfs_key key;
-	int index;
-	int err = 0;
 
 	if (objectid < BTRFS_FIRST_FREE_OBJECTID)
 		return ERR_PTR(-ESTALE);
@@ -75,25 +73,18 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = (u64)-1;
 
-	index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(root)) {
-		err = PTR_ERR(root);
-		goto fail;
-	}
+	root = btrfs_get_fs_root(fs_info, &key, true);
+	if (IS_ERR(root))
+		return ERR_CAST(root);
 
 	key.objectid = objectid;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
 	inode = btrfs_iget(sb, &key, root);
-	if (IS_ERR(inode)) {
-		err = PTR_ERR(inode);
-		goto fail;
-	}
-
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
+	btrfs_put_root(root);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
 
 	if (check_generation && generation != inode->i_generation) {
 		iput(inode);
@@ -101,9 +92,6 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
 	}
 
 	return d_obtain_alias(inode);
-fail:
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
-	return ERR_PTR(err);
 }
 
 static struct dentry *btrfs_fh_to_parent(struct super_block *sb, struct fid *fh,
@@ -152,7 +140,7 @@ static struct dentry *btrfs_fh_to_dentry(struct super_block *sb, struct fid *fh,
 	return btrfs_get_dentry(sb, objectid, root_objectid, generation, 1);
 }
 
-static struct dentry *btrfs_get_parent(struct dentry *child)
+struct dentry *btrfs_get_parent(struct dentry *child)
 {
 	struct inode *dir = d_inode(child);
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);

fs/btrfs/export.h

@@ -18,4 +18,9 @@ struct btrfs_fid {
 	u64 parent_root_objectid;
 } __attribute__ ((packed));
 
+struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
+				u64 root_objectid, u32 generation,
+				int check_generation);
+struct dentry *btrfs_get_parent(struct dentry *child);
+
 #endif

fs/btrfs/extent-io-tree.h

@@ -36,13 +36,14 @@ struct io_failure_record;
 #define CHUNK_TRIMMED				EXTENT_DEFRAG
 
 enum {
-	IO_TREE_FS_INFO_FREED_EXTENTS0,
-	IO_TREE_FS_INFO_FREED_EXTENTS1,
+	IO_TREE_FS_PINNED_EXTENTS,
+	IO_TREE_FS_EXCLUDED_EXTENTS,
 	IO_TREE_INODE_IO,
 	IO_TREE_INODE_IO_FAILURE,
 	IO_TREE_RELOC_BLOCKS,
 	IO_TREE_TRANS_DIRTY_PAGES,
 	IO_TREE_ROOT_DIRTY_LOG_PAGES,
+	IO_TREE_INODE_FILE_EXTENT,
 	IO_TREE_SELFTEST,
 };
 
@@ -222,6 +223,8 @@ int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
 			  struct extent_state **cached_state);
 void find_first_clear_extent_bit(struct extent_io_tree *tree, u64 start,
 				 u64 *start_ret, u64 *end_ret, unsigned bits);
+int find_contiguous_extent_bit(struct extent_io_tree *tree, u64 start,
+			       u64 *start_ret, u64 *end_ret, unsigned bits);
 int extent_invalidatepage(struct extent_io_tree *tree,
 			  struct page *page, unsigned long offset);
 bool btrfs_find_delalloc_range(struct extent_io_tree *tree, u64 *start,

fs/btrfs/extent_io.h

@@ -189,8 +189,8 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
 int try_release_extent_mapping(struct page *page, gfp_t mask);
 int try_release_extent_buffer(struct page *page);
 
-int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
-			  get_extent_t *get_extent, int mirror_num);
+int extent_read_full_page(struct page *page, get_extent_t *get_extent,
+			  int mirror_num);
 int extent_write_full_page(struct page *page, struct writeback_control *wbc);
 int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 			      int mode);
@@ -325,4 +325,11 @@ bool find_lock_delalloc_range(struct inode *inode,
 #endif
 struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 					       u64 start);
+
+#ifdef CONFIG_BTRFS_DEBUG
+void btrfs_extent_buffer_leak_debug_check(struct btrfs_fs_info *fs_info);
+#else
+#define btrfs_extent_buffer_leak_debug_check(fs_info)	do {} while (0)
+#endif
+
 #endif

fs/btrfs/file-item.c

@@ -23,6 +23,97 @@
 #define MAX_CSUM_ITEMS(r, size) (min_t(u32, __MAX_CSUM_ITEMS(r, size), \
 				       PAGE_SIZE))
 
+/**
+ * @inode - the inode we want to update the disk_i_size for
+ * @new_i_size - the i_size we want to set to, 0 if we use i_size
+ *
+ * With NO_HOLES set this simply sets the disk_is_size to whatever i_size_read()
+ * returns as it is perfectly fine with a file that has holes without hole file
+ * extent items.
+ *
+ * However without NO_HOLES we need to only return the area that is contiguous
+ * from the 0 offset of the file.  Otherwise we could end up adjust i_size up
+ * to an extent that has a gap in between.
+ *
+ * Finally new_i_size should only be set in the case of truncate where we're not
+ * ready to use i_size_read() as the limiter yet.
+ */
+void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size)
+{
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	u64 start, end, i_size;
+	int ret;
+
+	i_size = new_i_size ?: i_size_read(inode);
+	if (btrfs_fs_incompat(fs_info, NO_HOLES)) {
+		BTRFS_I(inode)->disk_i_size = i_size;
+		return;
+	}
+
+	spin_lock(&BTRFS_I(inode)->lock);
+	ret = find_contiguous_extent_bit(&BTRFS_I(inode)->file_extent_tree, 0,
+					 &start, &end, EXTENT_DIRTY);
+	if (!ret && start == 0)
+		i_size = min(i_size, end + 1);
+	else
+		i_size = 0;
+	BTRFS_I(inode)->disk_i_size = i_size;
+	spin_unlock(&BTRFS_I(inode)->lock);
+}
+
+/**
+ * @inode - the inode we're modifying
+ * @start - the start file offset of the file extent we've inserted
+ * @len - the logical length of the file extent item
+ *
+ * Call when we are inserting a new file extent where there was none before.
+ * Does not need to call this in the case where we're replacing an existing file
+ * extent, however if not sure it's fine to call this multiple times.
+ *
+ * The start and len must match the file extent item, so thus must be sectorsize
+ * aligned.
+ */
+int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start,
+				      u64 len)
+{
+	if (len == 0)
+		return 0;
+
+	ASSERT(IS_ALIGNED(start + len, inode->root->fs_info->sectorsize));
+
+	if (btrfs_fs_incompat(inode->root->fs_info, NO_HOLES))
+		return 0;
+	return set_extent_bits(&inode->file_extent_tree, start, start + len - 1,
+			       EXTENT_DIRTY);
+}
+
+/**
+ * @inode - the inode we're modifying
+ * @start - the start file offset of the file extent we've inserted
+ * @len - the logical length of the file extent item
+ *
+ * Called when we drop a file extent, for example when we truncate.  Doesn't
+ * need to be called for cases where we're replacing a file extent, like when
+ * we've COWed a file extent.
+ *
+ * The start and len must match the file extent item, so thus must be sectorsize
+ * aligned.
+ */
+int btrfs_inode_clear_file_extent_range(struct btrfs_inode *inode, u64 start,
+					u64 len)
+{
+	if (len == 0)
+		return 0;
+
+	ASSERT(IS_ALIGNED(start + len, inode->root->fs_info->sectorsize) ||
+	       len == (u64)-1);
+
+	if (btrfs_fs_incompat(inode->root->fs_info, NO_HOLES))
+		return 0;
+	return clear_extent_bit(&inode->file_extent_tree, start,
+				start + len - 1, EXTENT_DIRTY, 0, 0, NULL);
+}
+
 static inline u32 max_ordered_sum_bytes(struct btrfs_fs_info *fs_info,
 					u16 csum_size)
 {
@@ -949,18 +1040,7 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 
 	btrfs_item_key_to_cpu(leaf, &key, slot);
 	extent_start = key.offset;
-
-	if (type == BTRFS_FILE_EXTENT_REG ||
-	    type == BTRFS_FILE_EXTENT_PREALLOC) {
-		extent_end = extent_start +
-			btrfs_file_extent_num_bytes(leaf, fi);
-	} else if (type == BTRFS_FILE_EXTENT_INLINE) {
-		size_t size;
-		size = btrfs_file_extent_ram_bytes(leaf, fi);
-		extent_end = ALIGN(extent_start + size,
-				   fs_info->sectorsize);
-	}
-
+	extent_end = btrfs_file_extent_end(path);
 	em->ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
 	if (type == BTRFS_FILE_EXTENT_REG ||
 	    type == BTRFS_FILE_EXTENT_PREALLOC) {
@@ -1007,3 +1087,30 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 			  root->root_key.objectid);
 	}
 }
+
+/*
+ * Returns the end offset (non inclusive) of the file extent item the given path
+ * points to. If it points to an inline extent, the returned offset is rounded
+ * up to the sector size.
+ */
+u64 btrfs_file_extent_end(const struct btrfs_path *path)
+{
+	const struct extent_buffer *leaf = path->nodes[0];
+	const int slot = path->slots[0];
+	struct btrfs_file_extent_item *fi;
+	struct btrfs_key key;
+	u64 end;
+
+	btrfs_item_key_to_cpu(leaf, &key, slot);
+	ASSERT(key.type == BTRFS_EXTENT_DATA_KEY);
+	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+
+	if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) {
+		end = btrfs_file_extent_ram_bytes(leaf, fi);
+		end = ALIGN(key.offset + end, leaf->fs_info->sectorsize);
+	} else {
+		end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
+	}
+
+	return end;
+}

fs/btrfs/free-space-tree.c

@@ -1251,9 +1251,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
 	btrfs_free_tree_block(trans, free_space_root, free_space_root->node,
 			      0, 1);
 
-	free_extent_buffer(free_space_root->node);
-	free_extent_buffer(free_space_root->commit_root);
-	kfree(free_space_root);
+	btrfs_put_root(free_space_root);
 
 	return btrfs_commit_transaction(trans);
 

fs/btrfs/inode-map.c

@@ -515,7 +515,7 @@ int btrfs_save_ino_cache(struct btrfs_root *root,
 	trace_btrfs_space_reservation(fs_info, "ino_cache", trans->transid,
 				      trans->bytes_reserved, 0);
 	btrfs_block_rsv_release(fs_info, trans->block_rsv,
-				trans->bytes_reserved);
+				trans->bytes_reserved, NULL);
 out:
 	trans->block_rsv = rsv;
 	trans->bytes_reserved = num_bytes;

fs/btrfs/locking.h

@@ -6,6 +6,9 @@
 #ifndef BTRFS_LOCKING_H
 #define BTRFS_LOCKING_H
 
+#include <linux/atomic.h>
+#include <linux/wait.h>
+#include <linux/percpu_counter.h>
 #include "extent_io.h"
 
 #define BTRFS_WRITE_LOCK 1
@@ -13,6 +16,8 @@
 #define BTRFS_WRITE_LOCK_BLOCKING 3
 #define BTRFS_READ_LOCK_BLOCKING 4
 
+struct btrfs_path;
+
 void btrfs_tree_lock(struct extent_buffer *eb);
 void btrfs_tree_unlock(struct extent_buffer *eb);
 
@@ -48,4 +53,19 @@ static inline void btrfs_tree_unlock_rw(struct extent_buffer *eb, int rw)
 		BUG();
 }
 
+struct btrfs_drew_lock {
+	atomic_t readers;
+	struct percpu_counter writers;
+	wait_queue_head_t pending_writers;
+	wait_queue_head_t pending_readers;
+};
+
+int btrfs_drew_lock_init(struct btrfs_drew_lock *lock);
+void btrfs_drew_lock_destroy(struct btrfs_drew_lock *lock);
+void btrfs_drew_write_lock(struct btrfs_drew_lock *lock);
+bool btrfs_drew_try_write_lock(struct btrfs_drew_lock *lock);
+void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock);
+void btrfs_drew_read_lock(struct btrfs_drew_lock *lock);
+void btrfs_drew_read_unlock(struct btrfs_drew_lock *lock);
+
 #endif

fs/btrfs/props.c

@@ -383,7 +383,7 @@ static int inherit_props(struct btrfs_trans_handle *trans,
 
 		if (need_reserve) {
 			btrfs_block_rsv_release(fs_info, trans->block_rsv,
-					num_bytes);
+					num_bytes, NULL);
 			if (ret)
 				return ret;
 		}