Merge tag 'for-4.19-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "Mostly fixes and cleanups, nothing big, though the notable thing is
  the inserted/deleted lines delta -1124.

  User visible changes:
   - allow defrag on opened read-only files that have rw permissions;
     similar to what dedupe will allow on such files

  Core changes:
   - tree checker improvements, reported by fuzzing:
      * more checks for: block group items, essential trees
      * chunk type validation
      * mount time cross-checks that physical and logical chunks match
      * switch more error codes to EUCLEAN aka EFSCORRUPTED

  Fixes:
   - fsync corner case fixes

   - fix send failure when root has deleted files still open

   - send, fix incorrect file layout after hole punching beyond eof

   - fix races between mount and deice scan ioctl, found by fuzzing

   - fix deadlock when delayed iput is called from writeback on the same
     inode; rare but has been observed in practice, also removes code

   - fix pinned byte accounting, using the right percpu helpers; this
     should avoid some write IO inefficiency during low space conditions

   - don't remove block group that still has pinned bytes

   - reset on-disk device stats value after replace, otherwise this
     would report stale values for the new device

  Cleanups:
   - time64_t/timespec64 cleanups

   - remove remaining dead code in scrub handling NOCOW extents after
     disabling it in previous cycle

   - simplify fsync regarding ordered extents logic and remove all the
     related code

   - remove redundant arguments in order to reduce stack space
     consumption

   - remove support for V0 type of extents, not in use since 2.6.30

   - remove several unused structure members

   - fewer indirect function calls by inlining some callbacks

   - qgroup rescan timing fixes

   - vfs: iget cleanups"

* tag 'for-4.19-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (182 commits)
  btrfs: revert fs_devices state on error of btrfs_init_new_device
  btrfs: Exit gracefully when chunk map cannot be inserted to the tree
  btrfs: Introduce mount time chunk <-> dev extent mapping check
  btrfs: Verify that every chunk has corresponding block group at mount time
  btrfs: Check that each block group has corresponding chunk at mount time
  Btrfs: send, fix incorrect file layout after hole punching beyond eof
  btrfs: Use wrapper macro for rcu string to remove duplicate code
  btrfs: simplify btrfs_iget
  btrfs: lift make_bad_inode into btrfs_iget
  btrfs: simplify IS_ERR/PTR_ERR checks
  btrfs: btrfs_iget never returns an is_bad_inode inode
  btrfs: replace: Reset on-disk dev stats value after replace
  btrfs: extent-tree: Remove unused __btrfs_free_block_rsv
  btrfs: backref: Use ERR_CAST to return error code
  btrfs: Remove redundant btrfs_release_path from btrfs_unlink_subvol
  btrfs: Remove root parameter from btrfs_unlink_subvol
  btrfs: Remove fs_info from btrfs_add_root_ref
  btrfs: Remove fs_info from btrfs_del_root_ref
  btrfs: Remove fs_info from btrfs_del_root
  btrfs: Remove fs_info from btrfs_delete_delayed_dir_index
  ...

fs/btrfs/acl.c

@@ -30,23 +30,22 @@ struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 		name = XATTR_NAME_POSIX_ACL_DEFAULT;
 		break;
 	default:
-		BUG();
+		return ERR_PTR(-EINVAL);
 	}
 
-	size = btrfs_getxattr(inode, name, "", 0);
+	size = btrfs_getxattr(inode, name, NULL, 0);
 	if (size > 0) {
 		value = kzalloc(size, GFP_KERNEL);
 		if (!value)
 			return ERR_PTR(-ENOMEM);
 		size = btrfs_getxattr(inode, name, value, size);
 	}
-	if (size > 0) {
+	if (size > 0)
 		acl = posix_acl_from_xattr(&init_user_ns, value, size);
-	} else if (size == -ERANGE || size == -ENODATA || size == 0) {
+	else if (size == -ENODATA || size == 0)
 		acl = NULL;
-	} else {
-		acl = ERR_PTR(-EIO);
-	}
+	else
+		acl = ERR_PTR(size);
 	kfree(value);
 
 	return acl;

fs/btrfs/backref.c

@@ -925,7 +925,7 @@ static int add_inline_refs(const struct btrfs_fs_info *fs_info,
 		type = btrfs_get_extent_inline_ref_type(leaf, iref,
 							BTRFS_REF_TYPE_ANY);
 		if (type == BTRFS_REF_TYPE_INVALID)
-			return -EINVAL;
+			return -EUCLEAN;
 
 		offset = btrfs_extent_inline_ref_offset(leaf, iref);
 
@@ -1793,7 +1793,7 @@ static int get_extent_inline_ref(unsigned long *ptr,
 	*out_type = btrfs_get_extent_inline_ref_type(eb, *out_eiref,
 						     BTRFS_REF_TYPE_ANY);
 	if (*out_type == BTRFS_REF_TYPE_INVALID)
-		return -EINVAL;
+		return -EUCLEAN;
 
 	*ptr += btrfs_extent_inline_ref_size(*out_type);
 	WARN_ON(*ptr > end);
@@ -2225,7 +2225,7 @@ struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root,
 
 	fspath = init_data_container(total_bytes);
 	if (IS_ERR(fspath))
-		return (void *)fspath;
+		return ERR_CAST(fspath);
 
 	ifp = kmalloc(sizeof(*ifp), GFP_KERNEL);
 	if (!ifp) {

fs/btrfs/btrfs_inode.h

@@ -178,7 +178,7 @@ struct btrfs_inode {
 	struct btrfs_delayed_node *delayed_node;
 
 	/* File creation time. */
-	struct timespec i_otime;
+	struct timespec64 i_otime;
 
 	/* Hook into fs_info->delayed_iputs */
 	struct list_head delayed_iput;

fs/btrfs/check-integrity.c

@@ -1539,7 +1539,12 @@ static int btrfsic_map_block(struct btrfsic_state *state, u64 bytenr, u32 len,
 	}
 
 	device = multi->stripes[0].dev;
-	block_ctx_out->dev = btrfsic_dev_state_lookup(device->bdev->bd_dev);
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state) ||
+	    !device->bdev || !device->name)
+		block_ctx_out->dev = NULL;
+	else
+		block_ctx_out->dev = btrfsic_dev_state_lookup(
+							device->bdev->bd_dev);
 	block_ctx_out->dev_bytenr = multi->stripes[0].physical;
 	block_ctx_out->start = bytenr;
 	block_ctx_out->len = len;
@@ -1624,7 +1629,7 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 		bio = btrfs_io_bio_alloc(num_pages - i);
 		bio_set_dev(bio, block_ctx->dev->bdev);
 		bio->bi_iter.bi_sector = dev_bytenr >> 9;
-		bio_set_op_attrs(bio, REQ_OP_READ, 0);
+		bio->bi_opf = REQ_OP_READ;
 
 		for (j = i; j < num_pages; j++) {
 			ret = bio_add_page(bio, block_ctx->pagev[j],

fs/btrfs/compression.c

@@ -5,7 +5,6 @@
 
 #include <linux/kernel.h>
 #include <linux/bio.h>
-#include <linux/buffer_head.h>
 #include <linux/file.h>
 #include <linux/fs.h>
 #include <linux/pagemap.h>
@@ -14,10 +13,7 @@
 #include <linux/init.h>
 #include <linux/string.h>
 #include <linux/backing-dev.h>
-#include <linux/mpage.h>
-#include <linux/swap.h>
 #include <linux/writeback.h>
-#include <linux/bit_spinlock.h>
 #include <linux/slab.h>
 #include <linux/sched/mm.h>
 #include <linux/log2.h>
@@ -303,7 +299,6 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 	struct bio *bio = NULL;
 	struct compressed_bio *cb;
 	unsigned long bytes_left;
-	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	int pg_index = 0;
 	struct page *page;
 	u64 first_byte = disk_start;
@@ -342,9 +337,7 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		page = compressed_pages[pg_index];
 		page->mapping = inode->i_mapping;
 		if (bio->bi_iter.bi_size)
-			submit = io_tree->ops->merge_bio_hook(page, 0,
-							   PAGE_SIZE,
-							   bio, 0);
+			submit = btrfs_merge_bio_hook(page, 0, PAGE_SIZE, bio, 0);
 
 		page->mapping = NULL;
 		if (submit || bio_add_page(bio, page, PAGE_SIZE, 0) <
@@ -613,7 +606,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	cb->len = bio->bi_iter.bi_size;
 
 	comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
-	bio_set_op_attrs (comp_bio, REQ_OP_READ, 0);
+	comp_bio->bi_opf = REQ_OP_READ;
 	comp_bio->bi_private = cb;
 	comp_bio->bi_end_io = end_compressed_bio_read;
 	refcount_set(&cb->pending_bios, 1);
@@ -626,9 +619,8 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		page->index = em_start >> PAGE_SHIFT;
 
 		if (comp_bio->bi_iter.bi_size)
-			submit = tree->ops->merge_bio_hook(page, 0,
-							PAGE_SIZE,
-							comp_bio, 0);
+			submit = btrfs_merge_bio_hook(page, 0, PAGE_SIZE,
+					comp_bio, 0);
 
 		page->mapping = NULL;
 		if (submit || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
@@ -660,7 +652,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 			}
 
 			comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
-			bio_set_op_attrs(comp_bio, REQ_OP_READ, 0);
+			comp_bio->bi_opf = REQ_OP_READ;
 			comp_bio->bi_private = cb;
 			comp_bio->bi_end_io = end_compressed_bio_read;
 

fs/btrfs/ctree.c

@@ -888,11 +888,7 @@ int btrfs_block_can_be_shared(struct btrfs_root *root,
 	     btrfs_root_last_snapshot(&root->root_item) ||
 	     btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)))
 		return 1;
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
-	    btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV)
-		return 1;
-#endif
+
 	return 0;
 }
 
@@ -3128,8 +3124,7 @@ again:
  * higher levels
  *
  */
-static void fixup_low_keys(struct btrfs_fs_info *fs_info,
-			   struct btrfs_path *path,
+static void fixup_low_keys(struct btrfs_path *path,
 			   struct btrfs_disk_key *key, int level)
 {
 	int i;
@@ -3181,7 +3176,7 @@ void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
 	btrfs_set_item_key(eb, &disk_key, slot);
 	btrfs_mark_buffer_dirty(eb);
 	if (slot == 0)
-		fixup_low_keys(fs_info, path, &disk_key, 1);
+		fixup_low_keys(path, &disk_key, 1);
 }
 
 /*
@@ -3359,17 +3354,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans,
 
 	root_add_used(root, fs_info->nodesize);
 
-	memzero_extent_buffer(c, 0, sizeof(struct btrfs_header));
 	btrfs_set_header_nritems(c, 1);
-	btrfs_set_header_level(c, level);
-	btrfs_set_header_bytenr(c, c->start);
-	btrfs_set_header_generation(c, trans->transid);
-	btrfs_set_header_backref_rev(c, BTRFS_MIXED_BACKREF_REV);
-	btrfs_set_header_owner(c, root->root_key.objectid);
-
-	write_extent_buffer_fsid(c, fs_info->fsid);
-	write_extent_buffer_chunk_tree_uuid(c, fs_info->chunk_tree_uuid);
-
 	btrfs_set_node_key(c, &lower_key, 0);
 	btrfs_set_node_blockptr(c, 0, lower->start);
 	lower_gen = btrfs_header_generation(lower);
@@ -3498,15 +3483,7 @@ static noinline int split_node(struct btrfs_trans_handle *trans,
 		return PTR_ERR(split);
 
 	root_add_used(root, fs_info->nodesize);
-
-	memzero_extent_buffer(split, 0, sizeof(struct btrfs_header));
-	btrfs_set_header_level(split, btrfs_header_level(c));
-	btrfs_set_header_bytenr(split, split->start);
-	btrfs_set_header_generation(split, trans->transid);
-	btrfs_set_header_backref_rev(split, BTRFS_MIXED_BACKREF_REV);
-	btrfs_set_header_owner(split, root->root_key.objectid);
-	write_extent_buffer_fsid(split, fs_info->fsid);
-	write_extent_buffer_chunk_tree_uuid(split, fs_info->chunk_tree_uuid);
+	ASSERT(btrfs_header_level(c) == level);
 
 	ret = tree_mod_log_eb_copy(fs_info, split, c, 0, mid, c_nritems - mid);
 	if (ret) {
@@ -3945,7 +3922,7 @@ static noinline int __push_leaf_left(struct btrfs_fs_info *fs_info,
 		clean_tree_block(fs_info, right);
 
 	btrfs_item_key(right, &disk_key, 0);
-	fixup_low_keys(fs_info, path, &disk_key, 1);
+	fixup_low_keys(path, &disk_key, 1);
 
 	/* then fixup the leaf pointer in the path */
 	if (path->slots[0] < push_items) {
@@ -4292,15 +4269,6 @@ again:
 
 	root_add_used(root, fs_info->nodesize);
 
-	memzero_extent_buffer(right, 0, sizeof(struct btrfs_header));
-	btrfs_set_header_bytenr(right, right->start);
-	btrfs_set_header_generation(right, trans->transid);
-	btrfs_set_header_backref_rev(right, BTRFS_MIXED_BACKREF_REV);
-	btrfs_set_header_owner(right, root->root_key.objectid);
-	btrfs_set_header_level(right, 0);
-	write_extent_buffer_fsid(right, fs_info->fsid);
-	write_extent_buffer_chunk_tree_uuid(right, fs_info->chunk_tree_uuid);
-
 	if (split == 0) {
 		if (mid <= slot) {
 			btrfs_set_header_nritems(right, 0);
@@ -4320,7 +4288,7 @@ again:
 			path->nodes[0] = right;
 			path->slots[0] = 0;
 			if (path->slots[1] == 0)
-				fixup_low_keys(fs_info, path, &disk_key, 1);
+				fixup_low_keys(path, &disk_key, 1);
 		}
 		/*
 		 * We create a new leaf 'right' for the required ins_len and
@@ -4642,7 +4610,7 @@ void btrfs_truncate_item(struct btrfs_fs_info *fs_info,
 		btrfs_set_disk_key_offset(&disk_key, offset + size_diff);
 		btrfs_set_item_key(leaf, &disk_key, slot);
 		if (slot == 0)
-			fixup_low_keys(fs_info, path, &disk_key, 1);
+			fixup_low_keys(path, &disk_key, 1);
 	}
 
 	item = btrfs_item_nr(slot);
@@ -4744,7 +4712,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 
 	if (path->slots[0] == 0) {
 		btrfs_cpu_key_to_disk(&disk_key, cpu_key);
-		fixup_low_keys(fs_info, path, &disk_key, 1);
+		fixup_low_keys(path, &disk_key, 1);
 	}
 	btrfs_unlock_up_safe(path, 1);
 
@@ -4886,7 +4854,6 @@ int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
 		    int level, int slot)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *parent = path->nodes[level];
 	u32 nritems;
 	int ret;
@@ -4919,7 +4886,7 @@ static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
 		struct btrfs_disk_key disk_key;
 
 		btrfs_node_key(parent, &disk_key, 0);
-		fixup_low_keys(fs_info, path, &disk_key, level + 1);
+		fixup_low_keys(path, &disk_key, level + 1);
 	}
 	btrfs_mark_buffer_dirty(parent);
 }
@@ -5022,7 +4989,7 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 			struct btrfs_disk_key disk_key;
 
 			btrfs_item_key(leaf, &disk_key, 0);
-			fixup_low_keys(fs_info, path, &disk_key, 1);
+			fixup_low_keys(path, &disk_key, 1);
 		}
 
 		/* delete the leaf if it is mostly empty */

fs/btrfs/ctree.h

@@ -55,8 +55,6 @@ struct btrfs_ordered_sum;
 
 #define BTRFS_OLDEST_GENERATION	0ULL
 
-#define BTRFS_COMPAT_EXTENT_TREE_V0
-
 /*
  * the max metadata block size.  This limit is somewhat artificial,
  * but the memmove costs go through the roof for larger blocks.
@@ -86,6 +84,14 @@ static const int btrfs_csum_sizes[] = { 4 };
 
 #define BTRFS_DIRTY_METADATA_THRESH	SZ_32M
 
+/*
+ * Use large batch size to reduce overhead of metadata updates.  On the reader
+ * side, we only read it when we are close to ENOSPC and the read overhead is
+ * mostly related to the number of CPUs, so it is OK to use arbitrary large
+ * value here.
+ */
+#define BTRFS_TOTAL_BYTES_PINNED_BATCH	SZ_128M
+
 #define BTRFS_MAX_EXTENT_SIZE SZ_128M
 
 
@@ -342,8 +348,8 @@ struct btrfs_path {
 					sizeof(struct btrfs_item))
 struct btrfs_dev_replace {
 	u64 replace_state;	/* see #define above */
-	u64 time_started;	/* seconds since 1-Jan-1970 */
-	u64 time_stopped;	/* seconds since 1-Jan-1970 */
+	time64_t time_started;	/* seconds since 1-Jan-1970 */
+	time64_t time_stopped;	/* seconds since 1-Jan-1970 */
 	atomic64_t num_write_errors;
 	atomic64_t num_uncorrectable_read_errors;
 
@@ -359,8 +365,6 @@ struct btrfs_dev_replace {
 	struct btrfs_device *srcdev;
 	struct btrfs_device *tgtdev;
 
-	pid_t lock_owner;
-	atomic_t nesting_level;
 	struct mutex lock_finishing_cancel_unmount;
 	rwlock_t lock;
 	atomic_t read_locks;
@@ -1213,7 +1217,6 @@ struct btrfs_root {
 	u64 defrag_trans_start;
 	struct btrfs_key defrag_progress;
 	struct btrfs_key defrag_max;
-	char *name;
 
 	/* the dirty list is only used by non-reference counted roots */
 	struct list_head dirty_list;
@@ -2428,32 +2431,6 @@ static inline u32 btrfs_file_extent_inline_item_len(
 	return btrfs_item_size(eb, e) - BTRFS_FILE_EXTENT_INLINE_DATA_START;
 }
 
-/* this returns the number of file bytes represented by the inline item.
- * If an item is compressed, this is the uncompressed size
- */
-static inline u32 btrfs_file_extent_inline_len(const struct extent_buffer *eb,
-					int slot,
-					const struct btrfs_file_extent_item *fi)
-{
-	struct btrfs_map_token token;
-
-	btrfs_init_map_token(&token);
-	/*
-	 * return the space used on disk if this item isn't
-	 * compressed or encoded
-	 */
-	if (btrfs_token_file_extent_compression(eb, fi, &token) == 0 &&
-	    btrfs_token_file_extent_encryption(eb, fi, &token) == 0 &&
-	    btrfs_token_file_extent_other_encoding(eb, fi, &token) == 0) {
-		return btrfs_file_extent_inline_item_len(eb,
-							 btrfs_item_nr(slot));
-	}
-
-	/* otherwise use the ram bytes field */
-	return btrfs_token_file_extent_ram_bytes(eb, fi, &token);
-}
-
-
 /* btrfs_dev_stats_item */
 static inline u64 btrfs_dev_stats_value(const struct extent_buffer *eb,
 					const struct btrfs_dev_stats_item *ptr,
@@ -2676,7 +2653,6 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 				     u64 offset, u64 ram_bytes,
 				     struct btrfs_key *ins);
 int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   u64 root_objectid, u64 owner, u64 offset,
 				   struct btrfs_key *ins);
 int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes, u64 num_bytes,
@@ -2716,15 +2692,14 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info);
 int btrfs_read_block_groups(struct btrfs_fs_info *info);
 int btrfs_can_relocate(struct btrfs_fs_info *fs_info, u64 bytenr);
 int btrfs_make_block_group(struct btrfs_trans_handle *trans,
-			   struct btrfs_fs_info *fs_info, u64 bytes_used,
-			   u64 type, u64 chunk_offset, u64 size);
+			   u64 bytes_used, u64 type, u64 chunk_offset,
+			   u64 size);
 void btrfs_add_raid_kobjects(struct btrfs_fs_info *fs_info);
 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,
-			     struct btrfs_fs_info *fs_info, u64 group_start,
-			     struct extent_map *em);
+			     u64 group_start, struct extent_map *em);
 void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info);
 void btrfs_get_block_group_trimming(struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *cache);
@@ -2786,7 +2761,6 @@ void btrfs_init_metadata_block_rsv(struct btrfs_fs_info *fs_info,
 				   unsigned short type);
 void btrfs_free_block_rsv(struct btrfs_fs_info *fs_info,
 			  struct btrfs_block_rsv *rsv);
-void __btrfs_free_block_rsv(struct btrfs_block_rsv *rsv);
 int btrfs_block_rsv_add(struct btrfs_root *root,
 			struct btrfs_block_rsv *block_rsv, u64 num_bytes,
 			enum btrfs_reserve_flush_enum flush);
@@ -2803,8 +2777,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_rsv *block_rsv,
 			     u64 num_bytes);
-int btrfs_inc_block_group_ro(struct btrfs_fs_info *fs_info,
-			     struct btrfs_block_group_cache *cache);
+int btrfs_inc_block_group_ro(struct btrfs_block_group_cache *cache);
 void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group_cache(struct btrfs_fs_info *info);
 u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo);
@@ -2812,8 +2785,7 @@ int btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info,
 				   u64 start, u64 end);
 int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 			 u64 num_bytes, u64 *actual_bytes);
-int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info, u64 type);
+int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans, u64 type);
 int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range);
 
 int btrfs_init_space_info(struct btrfs_fs_info *fs_info);
@@ -2822,10 +2794,10 @@ int btrfs_delayed_refs_qgroup_accounting(struct btrfs_trans_handle *trans,
 int btrfs_start_write_no_snapshotting(struct btrfs_root *root);
 void btrfs_end_write_no_snapshotting(struct btrfs_root *root);
 void btrfs_wait_for_snapshot_creation(struct btrfs_root *root);
-void check_system_chunk(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, const u64 type);
+void check_system_chunk(struct btrfs_trans_handle *trans, const u64 type);
 u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 		       u64 start, u64 end);
+void btrfs_mark_bg_unused(struct btrfs_block_group_cache *bg);
 
 /* ctree.c */
 int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
@@ -3011,16 +2983,14 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq);
 
 /* root-item.c */
-int btrfs_add_root_ref(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info,
-		       u64 root_id, u64 ref_id, u64 dirid, u64 sequence,
-		       const char *name, int name_len);
-int btrfs_del_root_ref(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info,
-		       u64 root_id, u64 ref_id, u64 dirid, u64 *sequence,
-		       const char *name, int name_len);
+int btrfs_add_root_ref(struct btrfs_trans_handle *trans, u64 root_id,
+		       u64 ref_id, u64 dirid, u64 sequence, const char *name,
+		       int name_len);
+int btrfs_del_root_ref(struct btrfs_trans_handle *trans, u64 root_id,
+		       u64 ref_id, u64 dirid, u64 *sequence, const char *name,
+		       int name_len);
 int btrfs_del_root(struct btrfs_trans_handle *trans,
-		   struct btrfs_fs_info *fs_info, const struct btrfs_key *key);
+		   const struct btrfs_key *key);
 int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		      const struct btrfs_key *key,
 		      struct btrfs_root_item *item);
@@ -3196,7 +3166,7 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
 int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
 			 size_t size, struct bio *bio,
 			 unsigned long bio_flags);
-void btrfs_set_range_writeback(void *private_data, u64 start, u64 end);
+void btrfs_set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end);
 vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf);
 int btrfs_readpage(struct file *file, struct page *page);
 void btrfs_evict_inode(struct inode *inode);
@@ -3452,7 +3422,7 @@ do {								\
 #ifdef CONFIG_BTRFS_ASSERT
 
 __cold
-static inline void assfail(char *expr, char *file, int line)
+static inline void assfail(const char *expr, const char *file, int line)
 {
 	pr_err("assertion failed: %s, file: %s, line: %d\n",
 	       expr, file, line);
@@ -3465,6 +3435,13 @@ static inline void assfail(char *expr, char *file, int line)
 #define ASSERT(expr)	((void)0)
 #endif
 
+__cold
+static inline void btrfs_print_v0_err(struct btrfs_fs_info *fs_info)
+{
+	btrfs_err(fs_info,
+"Unsupported V0 extent filesystem detected. Aborting. Please re-create your filesystem with a newer kernel");
+}
+
 __printf(5, 6)
 __cold
 void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function,

fs/btrfs/delayed-inode.c

@@ -1222,7 +1222,7 @@ int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
 
 int btrfs_commit_inode_delayed_inode(struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_trans_handle *trans;
 	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
 	struct btrfs_path *path;
@@ -1418,7 +1418,6 @@ void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info)
 
 /* Will return 0 or -ENOMEM */
 int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   const char *name, int name_len,
 				   struct btrfs_inode *dir,
 				   struct btrfs_disk_key *disk_key, u8 type,
@@ -1458,11 +1457,10 @@ int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
 	 */
 	BUG_ON(ret);
 
-
 	mutex_lock(&delayed_node->mutex);
 	ret = __btrfs_add_delayed_insertion_item(delayed_node, delayed_item);
 	if (unlikely(ret)) {
-		btrfs_err(fs_info,
+		btrfs_err(trans->fs_info,
 			  "err add delayed dir index item(name: %.*s) into the insertion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
 			  name_len, name, delayed_node->root->objectid,
 			  delayed_node->inode_id, ret);
@@ -1495,7 +1493,6 @@ static int btrfs_delete_delayed_insertion_item(struct btrfs_fs_info *fs_info,
 }
 
 int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   struct btrfs_inode *dir, u64 index)
 {
 	struct btrfs_delayed_node *node;
@@ -1511,7 +1508,8 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 	item_key.type = BTRFS_DIR_INDEX_KEY;
 	item_key.offset = index;
 
-	ret = btrfs_delete_delayed_insertion_item(fs_info, node, &item_key);
+	ret = btrfs_delete_delayed_insertion_item(trans->fs_info, node,
+						  &item_key);
 	if (!ret)
 		goto end;
 
@@ -1533,7 +1531,7 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 	mutex_lock(&node->mutex);
 	ret = __btrfs_add_delayed_deletion_item(node, item);
 	if (unlikely(ret)) {
-		btrfs_err(fs_info,
+		btrfs_err(trans->fs_info,
 			  "err add delayed dir index item(index: %llu) into the deletion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
 			  index, node->root->objectid, node->inode_id, ret);
 		BUG();
@@ -1837,7 +1835,7 @@ release_node:
 
 int btrfs_delayed_delete_inode_ref(struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_delayed_node *delayed_node;
 
 	/*

fs/btrfs/delayed-inode.h

@@ -86,14 +86,12 @@ static inline void btrfs_init_delayed_root(
 }
 
 int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   const char *name, int name_len,
 				   struct btrfs_inode *dir,
 				   struct btrfs_disk_key *disk_key, u8 type,
 				   u64 index);
 
 int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   struct btrfs_inode *dir, u64 index);
 
 int btrfs_inode_delayed_dir_index_count(struct btrfs_inode *inode);

fs/btrfs/delayed-ref.c

@@ -709,13 +709,13 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info,
  * to make sure the delayed ref is eventually processed before this
  * transaction commits.
  */
-int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
-			       struct btrfs_trans_handle *trans,
+int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes, u64 parent,
 			       u64 ref_root,  int level, int action,
 			       struct btrfs_delayed_extent_op *extent_op,
 			       int *old_ref_mod, int *new_ref_mod)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_tree_ref *ref;
 	struct btrfs_delayed_ref_head *head_ref;
 	struct btrfs_delayed_ref_root *delayed_refs;
@@ -730,27 +730,33 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 	if (!ref)
 		return -ENOMEM;
 
+	head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
+	if (!head_ref) {
+		kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
+		return -ENOMEM;
+	}
+
+	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) &&
+	    is_fstree(ref_root)) {
+		record = kmalloc(sizeof(*record), GFP_NOFS);
+		if (!record) {
+			kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
+			kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref);
+			return -ENOMEM;
+		}
+	}
+
 	if (parent)
 		ref_type = BTRFS_SHARED_BLOCK_REF_KEY;
 	else
 		ref_type = BTRFS_TREE_BLOCK_REF_KEY;
+
 	init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes,
 				ref_root, action, ref_type);
 	ref->root = ref_root;
 	ref->parent = parent;
 	ref->level = level;
 
-	head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
-	if (!head_ref)
-		goto free_ref;
-
-	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) &&
-	    is_fstree(ref_root)) {
-		record = kmalloc(sizeof(*record), GFP_NOFS);
-		if (!record)
-			goto free_head_ref;
-	}
-
 	init_delayed_ref_head(head_ref, record, bytenr, num_bytes,
 			      ref_root, 0, action, false, is_system);
 	head_ref->extent_op = extent_op;
@@ -779,25 +785,18 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 		btrfs_qgroup_trace_extent_post(fs_info, record);
 
 	return 0;
-
-free_head_ref:
-	kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref);
-free_ref:
-	kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
-
-	return -ENOMEM;
 }
 
 /*
  * add a delayed data ref. it's similar to btrfs_add_delayed_tree_ref.
  */
-int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
-			       struct btrfs_trans_handle *trans,
+int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes,
 			       u64 parent, u64 ref_root,
 			       u64 owner, u64 offset, u64 reserved, int action,
 			       int *old_ref_mod, int *new_ref_mod)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_data_ref *ref;
 	struct btrfs_delayed_ref_head *head_ref;
 	struct btrfs_delayed_ref_root *delayed_refs;

fs/btrfs/delayed-ref.h

@@ -234,14 +234,12 @@ static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *hea
 		kmem_cache_free(btrfs_delayed_ref_head_cachep, head);
 }
 
-int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
-			       struct btrfs_trans_handle *trans,
+int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes, u64 parent,
 			       u64 ref_root, int level, int action,
 			       struct btrfs_delayed_extent_op *extent_op,
 			       int *old_ref_mod, int *new_ref_mod);
-int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
-			       struct btrfs_trans_handle *trans,
+int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes,
 			       u64 parent, u64 ref_root,
 			       u64 owner, u64 offset, u64 reserved, int action,

fs/btrfs/dev-replace.c

@@ -6,14 +6,9 @@
 #include <linux/sched.h>
 #include <linux/bio.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
 #include <linux/blkdev.h>
-#include <linux/random.h>
-#include <linux/iocontext.h>
-#include <linux/capability.h>
 #include <linux/kthread.h>
 #include <linux/math64.h>
-#include <asm/div64.h>
 #include "ctree.h"
 #include "extent_map.h"
 #include "disk-io.h"
@@ -465,7 +460,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	 * go to the tgtdev as well (refer to btrfs_map_block()).
 	 */
 	dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
-	dev_replace->time_started = get_seconds();
+	dev_replace->time_started = ktime_get_real_seconds();
 	dev_replace->cursor_left = 0;
 	dev_replace->committed_cursor_left = 0;
 	dev_replace->cursor_left_last_write_of_item = 0;
@@ -511,7 +506,7 @@ leave:
 	dev_replace->srcdev = NULL;
 	dev_replace->tgtdev = NULL;
 	btrfs_dev_replace_write_unlock(dev_replace);
-	btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
+	btrfs_destroy_dev_replace_tgtdev(tgt_device);
 	return ret;
 }
 
@@ -618,7 +613,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 			  : BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED;
 	dev_replace->tgtdev = NULL;
 	dev_replace->srcdev = NULL;
-	dev_replace->time_stopped = get_seconds();
+	dev_replace->time_stopped = ktime_get_real_seconds();
 	dev_replace->item_needs_writeback = 1;
 
 	/* replace old device with new one in mapping tree */
@@ -637,7 +632,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 		btrfs_rm_dev_replace_blocked(fs_info);
 		if (tgt_device)
-			btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
+			btrfs_destroy_dev_replace_tgtdev(tgt_device);
 		btrfs_rm_dev_replace_unblocked(fs_info);
 		mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
 
@@ -663,7 +658,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	tgt_device->commit_total_bytes = src_device->commit_total_bytes;
 	tgt_device->commit_bytes_used = src_device->bytes_used;
 
-	btrfs_assign_next_active_device(fs_info, src_device, tgt_device);
+	btrfs_assign_next_active_device(src_device, tgt_device);
 
 	list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
 	fs_info->fs_devices->rw_devices++;
@@ -672,10 +667,16 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 
 	btrfs_rm_dev_replace_blocked(fs_info);
 
-	btrfs_rm_dev_replace_remove_srcdev(fs_info, src_device);
+	btrfs_rm_dev_replace_remove_srcdev(src_device);
 
 	btrfs_rm_dev_replace_unblocked(fs_info);
 
+	/*
+	 * Increment dev_stats_ccnt so that btrfs_run_dev_stats() will
+	 * update on-disk dev stats value during commit transaction
+	 */
+	atomic_inc(&tgt_device->dev_stats_ccnt);
+
 	/*
 	 * this is again a consistent state where no dev_replace procedure
 	 * is running, the target device is part of the filesystem, the
@@ -807,7 +808,7 @@ int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
 		break;
 	}
 	dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED;
-	dev_replace->time_stopped = get_seconds();
+	dev_replace->time_stopped = ktime_get_real_seconds();
 	dev_replace->item_needs_writeback = 1;
 	btrfs_dev_replace_write_unlock(dev_replace);
 	btrfs_scrub_cancel(fs_info);
@@ -826,7 +827,7 @@ int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
 		btrfs_dev_name(tgt_device));
 
 	if (tgt_device)
-		btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
+		btrfs_destroy_dev_replace_tgtdev(tgt_device);
 
 leave:
 	mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
@@ -848,7 +849,7 @@ void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info)
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
 		dev_replace->replace_state =
 			BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
-		dev_replace->time_stopped = get_seconds();
+		dev_replace->time_stopped = ktime_get_real_seconds();
 		dev_replace->item_needs_writeback = 1;
 		btrfs_info(fs_info, "suspending dev_replace for unmount");
 		break;

fs/btrfs/dir-item.c

@@ -160,8 +160,8 @@ second_insert:
 	}
 	btrfs_release_path(path);
 
-	ret2 = btrfs_insert_delayed_dir_index(trans, root->fs_info, name,
-			name_len, dir, &disk_key, type, index);
+	ret2 = btrfs_insert_delayed_dir_index(trans, name, name_len, dir,
+					      &disk_key, type, index);
 out_free:
 	btrfs_free_path(path);
 	if (ret)

fs/btrfs/disk-io.c

@@ -5,8 +5,6 @@
 
 #include <linux/fs.h>
 #include <linux/blkdev.h>
-#include <linux/scatterlist.h>
-#include <linux/swap.h>
 #include <linux/radix-tree.h>
 #include <linux/writeback.h>
 #include <linux/buffer_head.h>
@@ -54,7 +52,6 @@
 
 static const struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
-static void free_fs_root(struct btrfs_root *root);
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
 static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 				      struct btrfs_fs_info *fs_info);
@@ -108,12 +105,9 @@ void __cold btrfs_end_io_wq_exit(void)
  */
 struct async_submit_bio {
 	void *private_data;
-	struct btrfs_fs_info *fs_info;
 	struct bio *bio;
 	extent_submit_bio_start_t *submit_bio_start;
-	extent_submit_bio_done_t *submit_bio_done;
 	int mirror_num;
-	unsigned long bio_flags;
 	/*
 	 * bio_offset is optional, can be used if the pages in the bio
 	 * can't tell us where in the file the bio should go
@@ -212,7 +206,7 @@ struct extent_map *btree_get_extent(struct btrfs_inode *inode,
 		struct page *page, size_t pg_offset, u64 start, u64 len,
 		int create)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct extent_map_tree *em_tree = &inode->extent_tree;
 	struct extent_map *em;
 	int ret;
@@ -615,8 +609,8 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 
 	found_start = btrfs_header_bytenr(eb);
 	if (found_start != eb->start) {
-		btrfs_err_rl(fs_info, "bad tree block start %llu %llu",
-			     found_start, eb->start);
+		btrfs_err_rl(fs_info, "bad tree block start, want %llu have %llu",
+			     eb->start, found_start);
 		ret = -EIO;
 		goto err;
 	}
@@ -628,8 +622,8 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 	}
 	found_level = btrfs_header_level(eb);
 	if (found_level >= BTRFS_MAX_LEVEL) {
-		btrfs_err(fs_info, "bad tree block level %d",
-			  (int)btrfs_header_level(eb));
+		btrfs_err(fs_info, "bad tree block level %d on %llu",
+			  (int)btrfs_header_level(eb), eb->start);
 		ret = -EIO;
 		goto err;
 	}
@@ -779,7 +773,7 @@ static void run_one_async_done(struct btrfs_work *work)
 		return;
 	}
 
-	async->submit_bio_done(async->private_data, async->bio, async->mirror_num);
+	btrfs_submit_bio_done(async->private_data, async->bio, async->mirror_num);
 }
 
 static void run_one_async_free(struct btrfs_work *work)
@@ -793,8 +787,7 @@ static void run_one_async_free(struct btrfs_work *work)
 blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags,
 				 u64 bio_offset, void *private_data,
-				 extent_submit_bio_start_t *submit_bio_start,
-				 extent_submit_bio_done_t *submit_bio_done)
+				 extent_submit_bio_start_t *submit_bio_start)
 {
 	struct async_submit_bio *async;
 
@@ -803,16 +796,13 @@ blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 		return BLK_STS_RESOURCE;
 
 	async->private_data = private_data;
-	async->fs_info = fs_info;
 	async->bio = bio;
 	async->mirror_num = mirror_num;
 	async->submit_bio_start = submit_bio_start;
-	async->submit_bio_done = submit_bio_done;
 
 	btrfs_init_work(&async->work, btrfs_worker_helper, run_one_async_start,
 			run_one_async_done, run_one_async_free);
 
-	async->bio_flags = bio_flags;
 	async->bio_offset = bio_offset;
 
 	async->status = 0;
@@ -851,24 +841,6 @@ static blk_status_t btree_submit_bio_start(void *private_data, struct bio *bio,
 	return btree_csum_one_bio(bio);
 }
 
-static blk_status_t btree_submit_bio_done(void *private_data, struct bio *bio,
-					    int mirror_num)
-{
-	struct inode *inode = private_data;
-	blk_status_t ret;
-
-	/*
-	 * when we're called for a write, we're already in the async
-	 * submission context.  Just jump into btrfs_map_bio
-	 */
-	ret = btrfs_map_bio(btrfs_sb(inode->i_sb), bio, mirror_num, 1);
-	if (ret) {
-		bio->bi_status = ret;
-		bio_endio(bio);
-	}
-	return ret;
-}
-
 static int check_async_write(struct btrfs_inode *bi)
 {
 	if (atomic_read(&bi->sync_writers))
@@ -911,8 +883,7 @@ static blk_status_t btree_submit_bio_hook(void *private_data, struct bio *bio,
 		 */
 		ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, 0,
 					  bio_offset, private_data,
-					  btree_submit_bio_start,
-					  btree_submit_bio_done);
+					  btree_submit_bio_start);
 	}
 
 	if (ret)
@@ -961,8 +932,9 @@ static int btree_writepages(struct address_space *mapping,
 
 		fs_info = BTRFS_I(mapping->host)->root->fs_info;
 		/* this is a bit racy, but that's ok */
-		ret = percpu_counter_compare(&fs_info->dirty_metadata_bytes,
-					     BTRFS_DIRTY_METADATA_THRESH);
+		ret = __percpu_counter_compare(&fs_info->dirty_metadata_bytes,
+					     BTRFS_DIRTY_METADATA_THRESH,
+					     fs_info->dirty_metadata_batch);
 		if (ret < 0)
 			return 0;
 	}
@@ -1181,7 +1153,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->highest_objectid = 0;
 	root->nr_delalloc_inodes = 0;
 	root->nr_ordered_extents = 0;
-	root->name = NULL;
 	root->inode_tree = RB_ROOT;
 	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
 	root->block_rsv = NULL;
@@ -1292,15 +1263,7 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 		goto fail;
 	}
 
-	memzero_extent_buffer(leaf, 0, sizeof(struct btrfs_header));
-	btrfs_set_header_bytenr(leaf, leaf->start);
-	btrfs_set_header_generation(leaf, trans->transid);
-	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
-	btrfs_set_header_owner(leaf, objectid);
 	root->node = leaf;
-
-	write_extent_buffer_fsid(leaf, fs_info->fsid);
-	write_extent_buffer_chunk_tree_uuid(leaf, fs_info->chunk_tree_uuid);
 	btrfs_mark_buffer_dirty(leaf);
 
 	root->commit_root = btrfs_root_node(root);
@@ -1374,14 +1337,8 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 		return ERR_CAST(leaf);
 	}
 
-	memzero_extent_buffer(leaf, 0, sizeof(struct btrfs_header));
-	btrfs_set_header_bytenr(leaf, leaf->start);
-	btrfs_set_header_generation(leaf, trans->transid);
-	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
-	btrfs_set_header_owner(leaf, BTRFS_TREE_LOG_OBJECTID);
 	root->node = leaf;
 
-	write_extent_buffer_fsid(root->node, fs_info->fsid);
 	btrfs_mark_buffer_dirty(root->node);
 	btrfs_tree_unlock(root->node);
 	return root;
@@ -1546,7 +1503,7 @@ int btrfs_init_fs_root(struct btrfs_root *root)
 
 	return 0;
 fail:
-	/* the caller is responsible to call free_fs_root */
+	/* The caller is responsible to call btrfs_free_fs_root */
 	return ret;
 }
 
@@ -1651,14 +1608,14 @@ again:
 	ret = btrfs_insert_fs_root(fs_info, root);
 	if (ret) {
 		if (ret == -EEXIST) {
-			free_fs_root(root);
+			btrfs_free_fs_root(root);
 			goto again;
 		}
 		goto fail;
 	}
 	return root;
 fail:
-	free_fs_root(root);
+	btrfs_free_fs_root(root);
 	return ERR_PTR(ret);
 }
 
@@ -1803,7 +1760,7 @@ static int transaction_kthread(void *arg)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_transaction *cur;
 	u64 transid;
-	unsigned long now;
+	time64_t now;
 	unsigned long delay;
 	bool cannot_commit;
 
@@ -1819,7 +1776,7 @@ static int transaction_kthread(void *arg)
 			goto sleep;
 		}
 
-		now = get_seconds();
+		now = ktime_get_seconds();
 		if (cur->state < TRANS_STATE_BLOCKED &&
 		    !test_bit(BTRFS_FS_NEED_ASYNC_COMMIT, &fs_info->flags) &&
 		    (now < cur->start_time ||
@@ -2196,8 +2153,6 @@ static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info)
 
 static void btrfs_init_dev_replace_locks(struct btrfs_fs_info *fs_info)
 {
-	fs_info->dev_replace.lock_owner = 0;
-	atomic_set(&fs_info->dev_replace.nesting_level, 0);
 	mutex_init(&fs_info->dev_replace.lock_finishing_cancel_unmount);
 	rwlock_init(&fs_info->dev_replace.lock);
 	atomic_set(&fs_info->dev_replace.read_locks, 0);
@@ -3075,6 +3030,13 @@ retry_root_backup:
 	fs_info->generation = generation;
 	fs_info->last_trans_committed = generation;
 
+	ret = btrfs_verify_dev_extents(fs_info);
+	if (ret) {
+		btrfs_err(fs_info,
+			  "failed to verify dev extents against chunks: %d",
+			  ret);
+		goto fail_block_groups;
+	}
 	ret = btrfs_recover_balance(fs_info);
 	if (ret) {
 		btrfs_err(fs_info, "failed to recover balance: %d", ret);
@@ -3875,10 +3837,10 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 		__btrfs_remove_free_space_cache(root->free_ino_pinned);
 	if (root->free_ino_ctl)
 		__btrfs_remove_free_space_cache(root->free_ino_ctl);
-	free_fs_root(root);
+	btrfs_free_fs_root(root);
 }
 
-static void free_fs_root(struct btrfs_root *root)
+void btrfs_free_fs_root(struct btrfs_root *root)
 {
 	iput(root->ino_cache_inode);
 	WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
@@ -3890,15 +3852,9 @@ static void free_fs_root(struct btrfs_root *root)
 	free_extent_buffer(root->commit_root);
 	kfree(root->free_ino_ctl);
 	kfree(root->free_ino_pinned);
-	kfree(root->name);
 	btrfs_put_fs_root(root);
 }
 
-void btrfs_free_fs_root(struct btrfs_root *root)
-{
-	free_fs_root(root);
-}
-
 int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 {
 	u64 root_objectid = 0;
@@ -4104,10 +4060,10 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	/*
 	 * This is a fast path so only do this check if we have sanity tests
-	 * enabled.  Normal people shouldn't be marking dummy buffers as dirty
+	 * enabled.  Normal people shouldn't be using umapped buffers as dirty
 	 * outside of the sanity tests.
 	 */
-	if (unlikely(test_bit(EXTENT_BUFFER_DUMMY, &buf->bflags)))
+	if (unlikely(test_bit(EXTENT_BUFFER_UNMAPPED, &buf->bflags)))
 		return;
 #endif
 	root = BTRFS_I(buf->pages[0]->mapping->host)->root;
@@ -4150,8 +4106,9 @@ static void __btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info,
 	if (flush_delayed)
 		btrfs_balance_delayed_items(fs_info);
 
-	ret = percpu_counter_compare(&fs_info->dirty_metadata_bytes,
-				     BTRFS_DIRTY_METADATA_THRESH);
+	ret = __percpu_counter_compare(&fs_info->dirty_metadata_bytes,
+				     BTRFS_DIRTY_METADATA_THRESH,
+				     fs_info->dirty_metadata_batch);
 	if (ret > 0) {
 		balance_dirty_pages_ratelimited(fs_info->btree_inode->i_mapping);
 	}
@@ -4563,21 +4520,11 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
 	return 0;
 }
 
-static struct btrfs_fs_info *btree_fs_info(void *private_data)
-{
-	struct inode *inode = private_data;
-	return btrfs_sb(inode->i_sb);
-}
-
 static const struct extent_io_ops btree_extent_io_ops = {
 	/* mandatory callbacks */
 	.submit_bio_hook = btree_submit_bio_hook,
 	.readpage_end_io_hook = btree_readpage_end_io_hook,
-	/* note we're sharing with inode.c for the merge bio hook */
-	.merge_bio_hook = btrfs_merge_bio_hook,
 	.readpage_io_failed_hook = btree_io_failed_hook,
-	.set_range_writeback = btrfs_set_range_writeback,
-	.tree_fs_info = btree_fs_info,
 
 	/* optional callbacks */
 };

fs/btrfs/disk-io.h

@@ -120,8 +120,9 @@ blk_status_t btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
 blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 			int mirror_num, unsigned long bio_flags,
 			u64 bio_offset, void *private_data,
-			extent_submit_bio_start_t *submit_bio_start,
-			extent_submit_bio_done_t *submit_bio_done);
+			extent_submit_bio_start_t *submit_bio_start);
+blk_status_t btrfs_submit_bio_done(void *private_data, struct bio *bio,
+			  int mirror_num);
 int btrfs_write_tree_block(struct extent_buffer *buf);
 void btrfs_wait_tree_block_writeback(struct extent_buffer *buf);
 int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,

fs/btrfs/extent-tree.c

@@ -52,24 +52,21 @@ enum {
 };
 
 static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
-			       struct btrfs_fs_info *fs_info,
-				struct btrfs_delayed_ref_node *node, u64 parent,
-				u64 root_objectid, u64 owner_objectid,
-				u64 owner_offset, int refs_to_drop,
-				struct btrfs_delayed_extent_op *extra_op);
+			       struct btrfs_delayed_ref_node *node, u64 parent,
+			       u64 root_objectid, u64 owner_objectid,
+			       u64 owner_offset, int refs_to_drop,
+			       struct btrfs_delayed_extent_op *extra_op);
 static void __run_delayed_extent_op(struct btrfs_delayed_extent_op *extent_op,
 				    struct extent_buffer *leaf,
 				    struct btrfs_extent_item *ei);
 static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
-				      struct btrfs_fs_info *fs_info,
 				      u64 parent, u64 root_objectid,
 				      u64 flags, u64 owner, u64 offset,
 				      struct btrfs_key *ins, int ref_mod);
 static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
 				     struct btrfs_delayed_ref_node *node,
 				     struct btrfs_delayed_extent_op *extent_op);
-static int do_chunk_alloc(struct btrfs_trans_handle *trans,
-			  struct btrfs_fs_info *fs_info, u64 flags,
+static int do_chunk_alloc(struct btrfs_trans_handle *trans, u64 flags,
 			  int force);
 static int find_next_key(struct btrfs_path *path, int level,
 			 struct btrfs_key *key);
@@ -220,9 +217,9 @@ static int add_excluded_extent(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-static void free_excluded_extents(struct btrfs_fs_info *fs_info,
-				  struct btrfs_block_group_cache *cache)
+static void free_excluded_extents(struct btrfs_block_group_cache *cache)
 {
+	struct btrfs_fs_info *fs_info = cache->fs_info;
 	u64 start, end;
 
 	start = cache->key.objectid;
@@ -234,9 +231,9 @@ static void free_excluded_extents(struct btrfs_fs_info *fs_info,
 			  start, end, EXTENT_UPTODATE);
 }
 
-static int exclude_super_stripes(struct btrfs_fs_info *fs_info,
-				 struct btrfs_block_group_cache *cache)
+static int exclude_super_stripes(struct btrfs_block_group_cache *cache)
 {
+	struct btrfs_fs_info *fs_info = cache->fs_info;
 	u64 bytenr;
 	u64 *logical;
 	int stripe_len;
@@ -558,7 +555,7 @@ static noinline void caching_thread(struct btrfs_work *work)
 	caching_ctl->progress = (u64)-1;
 
 	up_read(&fs_info->commit_root_sem);
-	free_excluded_extents(fs_info, block_group);
+	free_excluded_extents(block_group);
 	mutex_unlock(&caching_ctl->mutex);
 
 	wake_up(&caching_ctl->wait);
@@ -666,7 +663,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
 		wake_up(&caching_ctl->wait);
 		if (ret == 1) {
 			put_caching_control(caching_ctl);
-			free_excluded_extents(fs_info, cache);
+			free_excluded_extents(cache);
 			return 0;
 		}
 	} else {
@@ -758,7 +755,8 @@ static void add_pinned_bytes(struct btrfs_fs_info *fs_info, s64 num_bytes,
 
 	space_info = __find_space_info(fs_info, flags);
 	ASSERT(space_info);
-	percpu_counter_add(&space_info->total_bytes_pinned, num_bytes);
+	percpu_counter_add_batch(&space_info->total_bytes_pinned, num_bytes,
+		    BTRFS_TOTAL_BYTES_PINNED_BATCH);
 }
 
 /*
@@ -870,18 +868,16 @@ search_again:
 			num_refs = btrfs_extent_refs(leaf, ei);
 			extent_flags = btrfs_extent_flags(leaf, ei);
 		} else {
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-			struct btrfs_extent_item_v0 *ei0;
-			BUG_ON(item_size != sizeof(*ei0));
-			ei0 = btrfs_item_ptr(leaf, path->slots[0],
-					     struct btrfs_extent_item_v0);
-			num_refs = btrfs_extent_refs_v0(leaf, ei0);
-			/* FIXME: this isn't correct for data */
-			extent_flags = BTRFS_BLOCK_FLAG_FULL_BACKREF;
-#else
-			BUG();
-#endif
+			ret = -EINVAL;
+			btrfs_print_v0_err(fs_info);
+			if (trans)
+				btrfs_abort_transaction(trans, ret);
+			else
+				btrfs_handle_fs_error(fs_info, ret, NULL);
+
+			goto out_free;
 		}
+
 		BUG_ON(num_refs == 0);
 	} else {
 		num_refs = 0;
@@ -1039,89 +1035,6 @@ out_free:
  * tree block info structure.
  */
 
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-static int convert_extent_item_v0(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
-				  struct btrfs_path *path,
-				  u64 owner, u32 extra_size)
-{
-	struct btrfs_root *root = fs_info->extent_root;
-	struct btrfs_extent_item *item;
-	struct btrfs_extent_item_v0 *ei0;
-	struct btrfs_extent_ref_v0 *ref0;
-	struct btrfs_tree_block_info *bi;
-	struct extent_buffer *leaf;
-	struct btrfs_key key;
-	struct btrfs_key found_key;
-	u32 new_size = sizeof(*item);
-	u64 refs;
-	int ret;
-
-	leaf = path->nodes[0];
-	BUG_ON(btrfs_item_size_nr(leaf, path->slots[0]) != sizeof(*ei0));
-
-	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-	ei0 = btrfs_item_ptr(leaf, path->slots[0],
-			     struct btrfs_extent_item_v0);
-	refs = btrfs_extent_refs_v0(leaf, ei0);
-
-	if (owner == (u64)-1) {
-		while (1) {
-			if (path->slots[0] >= btrfs_header_nritems(leaf)) {
-				ret = btrfs_next_leaf(root, path);
-				if (ret < 0)
-					return ret;
-				BUG_ON(ret > 0); /* Corruption */
-				leaf = path->nodes[0];
-			}
-			btrfs_item_key_to_cpu(leaf, &found_key,
-					      path->slots[0]);
-			BUG_ON(key.objectid != found_key.objectid);
-			if (found_key.type != BTRFS_EXTENT_REF_V0_KEY) {
-				path->slots[0]++;
-				continue;
-			}
-			ref0 = btrfs_item_ptr(leaf, path->slots[0],
-					      struct btrfs_extent_ref_v0);
-			owner = btrfs_ref_objectid_v0(leaf, ref0);
-			break;
-		}
-	}
-	btrfs_release_path(path);
-
-	if (owner < BTRFS_FIRST_FREE_OBJECTID)
-		new_size += sizeof(*bi);
-
-	new_size -= sizeof(*ei0);
-	ret = btrfs_search_slot(trans, root, &key, path,
-				new_size + extra_size, 1);
-	if (ret < 0)
-		return ret;
-	BUG_ON(ret); /* Corruption */
-
-	btrfs_extend_item(fs_info, path, new_size);
-
-	leaf = path->nodes[0];
-	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
-	btrfs_set_extent_refs(leaf, item, refs);
-	/* FIXME: get real generation */
-	btrfs_set_extent_generation(leaf, item, 0);
-	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
-		btrfs_set_extent_flags(leaf, item,
-				       BTRFS_EXTENT_FLAG_TREE_BLOCK |
-				       BTRFS_BLOCK_FLAG_FULL_BACKREF);
-		bi = (struct btrfs_tree_block_info *)(item + 1);
-		/* FIXME: get first key of the block */
-		memzero_extent_buffer(leaf, (unsigned long)bi, sizeof(*bi));
-		btrfs_set_tree_block_level(leaf, bi, (int)owner);
-	} else {
-		btrfs_set_extent_flags(leaf, item, BTRFS_EXTENT_FLAG_DATA);
-	}
-	btrfs_mark_buffer_dirty(leaf);
-	return 0;
-}
-#endif
-
 /*
  * is_data == BTRFS_REF_TYPE_BLOCK, tree block type is required,
  * is_data == BTRFS_REF_TYPE_DATA, data type is requried,
@@ -1216,13 +1129,12 @@ static int match_extent_data_ref(struct extent_buffer *leaf,
 }
 
 static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans,
-					   struct btrfs_fs_info *fs_info,
 					   struct btrfs_path *path,
 					   u64 bytenr, u64 parent,
 					   u64 root_objectid,
 					   u64 owner, u64 offset)
 {
-	struct btrfs_root *root = fs_info->extent_root;
+	struct btrfs_root *root = trans->fs_info->extent_root;
 	struct btrfs_key key;
 	struct btrfs_extent_data_ref *ref;
 	struct extent_buffer *leaf;
@@ -1251,17 +1163,6 @@ again:
 	if (parent) {
 		if (!ret)
 			return 0;
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		key.type = BTRFS_EXTENT_REF_V0_KEY;
-		btrfs_release_path(path);
-		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-		if (ret < 0) {
-			err = ret;
-			goto fail;
-		}
-		if (!ret)
-			return 0;
-#endif
 		goto fail;
 	}
 
@@ -1304,13 +1205,12 @@ fail:
 }
 
 static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans,
-					   struct btrfs_fs_info *fs_info,
 					   struct btrfs_path *path,
 					   u64 bytenr, u64 parent,
 					   u64 root_objectid, u64 owner,
 					   u64 offset, int refs_to_add)
 {
-	struct btrfs_root *root = fs_info->extent_root;
+	struct btrfs_root *root = trans->fs_info->extent_root;
 	struct btrfs_key key;
 	struct extent_buffer *leaf;
 	u32 size;
@@ -1384,7 +1284,6 @@ fail:
 }
 
 static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans,
-					   struct btrfs_fs_info *fs_info,
 					   struct btrfs_path *path,
 					   int refs_to_drop, int *last_ref)
 {
@@ -1406,13 +1305,10 @@ static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans,
 		ref2 = btrfs_item_ptr(leaf, path->slots[0],
 				      struct btrfs_shared_data_ref);
 		num_refs = btrfs_shared_data_ref_count(leaf, ref2);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	} else if (key.type == BTRFS_EXTENT_REF_V0_KEY) {
-		struct btrfs_extent_ref_v0 *ref0;
-		ref0 = btrfs_item_ptr(leaf, path->slots[0],
-				      struct btrfs_extent_ref_v0);
-		num_refs = btrfs_ref_count_v0(leaf, ref0);
-#endif
+	} else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) {
+		btrfs_print_v0_err(trans->fs_info);
+		btrfs_abort_transaction(trans, -EINVAL);
+		return -EINVAL;
 	} else {
 		BUG();
 	}
@@ -1421,21 +1317,13 @@ static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans,
 	num_refs -= refs_to_drop;
 
 	if (num_refs == 0) {
-		ret = btrfs_del_item(trans, fs_info->extent_root, path);
+		ret = btrfs_del_item(trans, trans->fs_info->extent_root, path);
 		*last_ref = 1;
 	} else {
 		if (key.type == BTRFS_EXTENT_DATA_REF_KEY)
 			btrfs_set_extent_data_ref_count(leaf, ref1, num_refs);
 		else if (key.type == BTRFS_SHARED_DATA_REF_KEY)
 			btrfs_set_shared_data_ref_count(leaf, ref2, num_refs);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		else {
-			struct btrfs_extent_ref_v0 *ref0;
-			ref0 = btrfs_item_ptr(leaf, path->slots[0],
-					struct btrfs_extent_ref_v0);
-			btrfs_set_ref_count_v0(leaf, ref0, num_refs);
-		}
-#endif
 		btrfs_mark_buffer_dirty(leaf);
 	}
 	return ret;
@@ -1453,6 +1341,8 @@ static noinline u32 extent_data_ref_count(struct btrfs_path *path,
 
 	leaf = path->nodes[0];
 	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+
+	BUG_ON(key.type == BTRFS_EXTENT_REF_V0_KEY);
 	if (iref) {
 		/*
 		 * If type is invalid, we should have bailed out earlier than
@@ -1475,13 +1365,6 @@ static noinline u32 extent_data_ref_count(struct btrfs_path *path,
 		ref2 = btrfs_item_ptr(leaf, path->slots[0],
 				      struct btrfs_shared_data_ref);
 		num_refs = btrfs_shared_data_ref_count(leaf, ref2);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	} else if (key.type == BTRFS_EXTENT_REF_V0_KEY) {
-		struct btrfs_extent_ref_v0 *ref0;
-		ref0 = btrfs_item_ptr(leaf, path->slots[0],
-				      struct btrfs_extent_ref_v0);
-		num_refs = btrfs_ref_count_v0(leaf, ref0);
-#endif
 	} else {
 		WARN_ON(1);
 	}
@@ -1489,12 +1372,11 @@ static noinline u32 extent_data_ref_count(struct btrfs_path *path,
 }
 
 static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans,
-					  struct btrfs_fs_info *fs_info,
 					  struct btrfs_path *path,
 					  u64 bytenr, u64 parent,
 					  u64 root_objectid)
 {
-	struct btrfs_root *root = fs_info->extent_root;
+	struct btrfs_root *root = trans->fs_info->extent_root;
 	struct btrfs_key key;
 	int ret;
 
@@ -1510,20 +1392,10 @@ static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans,
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 	if (ret > 0)
 		ret = -ENOENT;
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (ret == -ENOENT && parent) {
-		btrfs_release_path(path);
-		key.type = BTRFS_EXTENT_REF_V0_KEY;
-		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-		if (ret > 0)
-			ret = -ENOENT;
-	}
-#endif
 	return ret;
 }
 
 static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans,
-					  struct btrfs_fs_info *fs_info,
 					  struct btrfs_path *path,
 					  u64 bytenr, u64 parent,
 					  u64 root_objectid)
@@ -1540,7 +1412,7 @@ static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans,
 		key.offset = root_objectid;
 	}
 
-	ret = btrfs_insert_empty_item(trans, fs_info->extent_root,
+	ret = btrfs_insert_empty_item(trans, trans->fs_info->extent_root,
 				      path, &key, 0);
 	btrfs_release_path(path);
 	return ret;
@@ -1599,13 +1471,13 @@ static int find_next_key(struct btrfs_path *path, int level,
  */
 static noinline_for_stack
 int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_path *path,
 				 struct btrfs_extent_inline_ref **ref_ret,
 				 u64 bytenr, u64 num_bytes,
 				 u64 parent, u64 root_objectid,
 				 u64 owner, u64 offset, int insert)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = fs_info->extent_root;
 	struct btrfs_key key;
 	struct extent_buffer *leaf;
@@ -1635,8 +1507,8 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
 		extra_size = -1;
 
 	/*
-	 * Owner is our parent level, so we can just add one to get the level
-	 * for the block we are interested in.
+	 * Owner is our level, so we can just add one to get the level for the
+	 * block we are interested in.
 	 */
 	if (skinny_metadata && owner < BTRFS_FIRST_FREE_OBJECTID) {
 		key.type = BTRFS_METADATA_ITEM_KEY;
@@ -1684,23 +1556,12 @@ again:
 
 	leaf = path->nodes[0];
 	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (item_size < sizeof(*ei)) {
-		if (!insert) {
-			err = -ENOENT;
-			goto out;
-		}
-		ret = convert_extent_item_v0(trans, fs_info, path, owner,
-					     extra_size);
-		if (ret < 0) {
-			err = ret;
-			goto out;
-		}
-		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	if (unlikely(item_size < sizeof(*ei))) {
+		err = -EINVAL;
+		btrfs_print_v0_err(fs_info);
+		btrfs_abort_transaction(trans, err);
+		goto out;
 	}
-#endif
-	BUG_ON(item_size < sizeof(*ei));
 
 	ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 	flags = btrfs_extent_flags(leaf, ei);
@@ -1727,7 +1588,7 @@ again:
 		iref = (struct btrfs_extent_inline_ref *)ptr;
 		type = btrfs_get_extent_inline_ref_type(leaf, iref, needed);
 		if (type == BTRFS_REF_TYPE_INVALID) {
-			err = -EINVAL;
+			err = -EUCLEAN;
 			goto out;
 		}
 
@@ -1863,7 +1724,6 @@ void setup_inline_extent_backref(struct btrfs_fs_info *fs_info,
 }
 
 static int lookup_extent_backref(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_path *path,
 				 struct btrfs_extent_inline_ref **ref_ret,
 				 u64 bytenr, u64 num_bytes, u64 parent,
@@ -1871,9 +1731,9 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
 {
 	int ret;
 
-	ret = lookup_inline_extent_backref(trans, fs_info, path, ref_ret,
-					   bytenr, num_bytes, parent,
-					   root_objectid, owner, offset, 0);
+	ret = lookup_inline_extent_backref(trans, path, ref_ret, bytenr,
+					   num_bytes, parent, root_objectid,
+					   owner, offset, 0);
 	if (ret != -ENOENT)
 		return ret;
 
@@ -1881,12 +1741,11 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
 	*ref_ret = NULL;
 
 	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
-		ret = lookup_tree_block_ref(trans, fs_info, path, bytenr,
-					    parent, root_objectid);
+		ret = lookup_tree_block_ref(trans, path, bytenr, parent,
+					    root_objectid);
 	} else {
-		ret = lookup_extent_data_ref(trans, fs_info, path, bytenr,
-					     parent, root_objectid, owner,
-					     offset);
+		ret = lookup_extent_data_ref(trans, path, bytenr, parent,
+					     root_objectid, owner, offset);
 	}
 	return ret;
 }
@@ -1895,14 +1754,14 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
  * helper to update/remove inline back ref
  */
 static noinline_for_stack
-void update_inline_extent_backref(struct btrfs_fs_info *fs_info,
-				  struct btrfs_path *path,
+void update_inline_extent_backref(struct btrfs_path *path,
 				  struct btrfs_extent_inline_ref *iref,
 				  int refs_to_mod,
 				  struct btrfs_delayed_extent_op *extent_op,
 				  int *last_ref)
 {
-	struct extent_buffer *leaf;
+	struct extent_buffer *leaf = path->nodes[0];
+	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	struct btrfs_extent_item *ei;
 	struct btrfs_extent_data_ref *dref = NULL;
 	struct btrfs_shared_data_ref *sref = NULL;
@@ -1913,7 +1772,6 @@ void update_inline_extent_backref(struct btrfs_fs_info *fs_info,
 	int type;
 	u64 refs;
 
-	leaf = path->nodes[0];
 	ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 	refs = btrfs_extent_refs(leaf, ei);
 	WARN_ON(refs_to_mod < 0 && refs + refs_to_mod <= 0);
@@ -1965,7 +1823,6 @@ void update_inline_extent_backref(struct btrfs_fs_info *fs_info,
 
 static noinline_for_stack
 int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_path *path,
 				 u64 bytenr, u64 num_bytes, u64 parent,
 				 u64 root_objectid, u64 owner,
@@ -1975,15 +1832,15 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
 	struct btrfs_extent_inline_ref *iref;
 	int ret;
 
-	ret = lookup_inline_extent_backref(trans, fs_info, path, &iref,
-					   bytenr, num_bytes, parent,
-					   root_objectid, owner, offset, 1);
+	ret = lookup_inline_extent_backref(trans, path, &iref, bytenr,
+					   num_bytes, parent, root_objectid,
+					   owner, offset, 1);
 	if (ret == 0) {
 		BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID);
-		update_inline_extent_backref(fs_info, path, iref,
-					     refs_to_add, extent_op, NULL);
+		update_inline_extent_backref(path, iref, refs_to_add,
+					     extent_op, NULL);
 	} else if (ret == -ENOENT) {
-		setup_inline_extent_backref(fs_info, path, iref, parent,
+		setup_inline_extent_backref(trans->fs_info, path, iref, parent,
 					    root_objectid, owner, offset,
 					    refs_to_add, extent_op);
 		ret = 0;
@@ -1992,7 +1849,6 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
 }
 
 static int insert_extent_backref(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_path *path,
 				 u64 bytenr, u64 parent, u64 root_objectid,
 				 u64 owner, u64 offset, int refs_to_add)
@@ -2000,18 +1856,17 @@ static int insert_extent_backref(struct btrfs_trans_handle *trans,
 	int ret;
 	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
 		BUG_ON(refs_to_add != 1);
-		ret = insert_tree_block_ref(trans, fs_info, path, bytenr,
-					    parent, root_objectid);
+		ret = insert_tree_block_ref(trans, path, bytenr, parent,
+					    root_objectid);
 	} else {
-		ret = insert_extent_data_ref(trans, fs_info, path, bytenr,
-					     parent, root_objectid,
-					     owner, offset, refs_to_add);
+		ret = insert_extent_data_ref(trans, path, bytenr, parent,
+					     root_objectid, owner, offset,
+					     refs_to_add);
 	}
 	return ret;
 }
 
 static int remove_extent_backref(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_path *path,
 				 struct btrfs_extent_inline_ref *iref,
 				 int refs_to_drop, int is_data, int *last_ref)
@@ -2020,14 +1875,14 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans,
 
 	BUG_ON(!is_data && refs_to_drop != 1);
 	if (iref) {
-		update_inline_extent_backref(fs_info, path, iref,
-					     -refs_to_drop, NULL, last_ref);
+		update_inline_extent_backref(path, iref, -refs_to_drop, NULL,
+					     last_ref);
 	} else if (is_data) {
-		ret = remove_extent_data_ref(trans, fs_info, path, refs_to_drop,
+		ret = remove_extent_data_ref(trans, path, refs_to_drop,
 					     last_ref);
 	} else {
 		*last_ref = 1;
-		ret = btrfs_del_item(trans, fs_info->extent_root, path);
+		ret = btrfs_del_item(trans, trans->fs_info->extent_root, path);
 	}
 	return ret;
 }
@@ -2185,13 +2040,13 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 			   owner, offset, BTRFS_ADD_DELAYED_REF);
 
 	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
-		ret = btrfs_add_delayed_tree_ref(fs_info, trans, bytenr,
+		ret = btrfs_add_delayed_tree_ref(trans, bytenr,
 						 num_bytes, parent,
 						 root_objectid, (int)owner,
 						 BTRFS_ADD_DELAYED_REF, NULL,
 						 &old_ref_mod, &new_ref_mod);
 	} else {
-		ret = btrfs_add_delayed_data_ref(fs_info, trans, bytenr,
+		ret = btrfs_add_delayed_data_ref(trans, bytenr,
 						 num_bytes, parent,
 						 root_objectid, owner, offset,
 						 0, BTRFS_ADD_DELAYED_REF,
@@ -2207,8 +2062,41 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
+/*
+ * __btrfs_inc_extent_ref - insert backreference for a given extent
+ *
+ * @trans:	    Handle of transaction
+ *
+ * @node:	    The delayed ref node used to get the bytenr/length for
+ *		    extent whose references are incremented.
+ *
+ * @parent:	    If this is a shared extent (BTRFS_SHARED_DATA_REF_KEY/
+ *		    BTRFS_SHARED_BLOCK_REF_KEY) then it holds the logical
+ *		    bytenr of the parent block. Since new extents are always
+ *		    created with indirect references, this will only be the case
+ *		    when relocating a shared extent. In that case, root_objectid
+ *		    will be BTRFS_TREE_RELOC_OBJECTID. Otheriwse, parent must
+ *		    be 0
+ *
+ * @root_objectid:  The id of the root where this modification has originated,
+ *		    this can be either one of the well-known metadata trees or
+ *		    the subvolume id which references this extent.
+ *
+ * @owner:	    For data extents it is the inode number of the owning file.
+ *		    For metadata extents this parameter holds the level in the
+ *		    tree of the extent.
+ *
+ * @offset:	    For metadata extents the offset is ignored and is currently
+ *		    always passed as 0. For data extents it is the fileoffset
+ *		    this extent belongs to.
+ *
+ * @refs_to_add     Number of references to add
+ *
+ * @extent_op       Pointer to a structure, holding information necessary when
+ *                  updating a tree block's flags
+ *
+ */
 static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_delayed_ref_node *node,
 				  u64 parent, u64 root_objectid,
 				  u64 owner, u64 offset, int refs_to_add,
@@ -2230,10 +2118,9 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	path->reada = READA_FORWARD;
 	path->leave_spinning = 1;
 	/* this will setup the path even if it fails to insert the back ref */
-	ret = insert_inline_extent_backref(trans, fs_info, path, bytenr,
-					   num_bytes, parent, root_objectid,
-					   owner, offset,
-					   refs_to_add, extent_op);
+	ret = insert_inline_extent_backref(trans, path, bytenr, num_bytes,
+					   parent, root_objectid, owner,
+					   offset, refs_to_add, extent_op);
 	if ((ret < 0 && ret != -EAGAIN) || !ret)
 		goto out;
 
@@ -2256,8 +2143,8 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	path->reada = READA_FORWARD;
 	path->leave_spinning = 1;
 	/* now insert the actual backref */
-	ret = insert_extent_backref(trans, fs_info, path, bytenr, parent,
-				    root_objectid, owner, offset, refs_to_add);
+	ret = insert_extent_backref(trans, path, bytenr, parent, root_objectid,
+				    owner, offset, refs_to_add);
 	if (ret)
 		btrfs_abort_transaction(trans, ret);
 out:
@@ -2266,7 +2153,6 @@ out:
 }
 
 static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
 				struct btrfs_delayed_ref_node *node,
 				struct btrfs_delayed_extent_op *extent_op,
 				int insert_reserved)
@@ -2283,7 +2169,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
 	ins.type = BTRFS_EXTENT_ITEM_KEY;
 
 	ref = btrfs_delayed_node_to_data_ref(node);
-	trace_run_delayed_data_ref(fs_info, node, ref, node->action);
+	trace_run_delayed_data_ref(trans->fs_info, node, ref, node->action);
 
 	if (node->type == BTRFS_SHARED_DATA_REF_KEY)
 		parent = ref->parent;
@@ -2292,17 +2178,16 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
 	if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
 		if (extent_op)
 			flags |= extent_op->flags_to_set;
-		ret = alloc_reserved_file_extent(trans, fs_info,
-						 parent, ref_root, flags,
-						 ref->objectid, ref->offset,
-						 &ins, node->ref_mod);
+		ret = alloc_reserved_file_extent(trans, parent, ref_root,
+						 flags, ref->objectid,
+						 ref->offset, &ins,
+						 node->ref_mod);
 	} else if (node->action == BTRFS_ADD_DELAYED_REF) {
-		ret = __btrfs_inc_extent_ref(trans, fs_info, node, parent,
-					     ref_root, ref->objectid,
-					     ref->offset, node->ref_mod,
-					     extent_op);
+		ret = __btrfs_inc_extent_ref(trans, node, parent, ref_root,
+					     ref->objectid, ref->offset,
+					     node->ref_mod, extent_op);
 	} else if (node->action == BTRFS_DROP_DELAYED_REF) {
-		ret = __btrfs_free_extent(trans, fs_info, node, parent,
+		ret = __btrfs_free_extent(trans, node, parent,
 					  ref_root, ref->objectid,
 					  ref->offset, node->ref_mod,
 					  extent_op);
@@ -2331,10 +2216,10 @@ static void __run_delayed_extent_op(struct btrfs_delayed_extent_op *extent_op,
 }
 
 static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_delayed_ref_head *head,
 				 struct btrfs_delayed_extent_op *extent_op)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_key key;
 	struct btrfs_path *path;
 	struct btrfs_extent_item *ei;
@@ -2400,18 +2285,14 @@ again:
 
 	leaf = path->nodes[0];
 	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (item_size < sizeof(*ei)) {
-		ret = convert_extent_item_v0(trans, fs_info, path, (u64)-1, 0);
-		if (ret < 0) {
-			err = ret;
-			goto out;
-		}
-		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+
+	if (unlikely(item_size < sizeof(*ei))) {
+		err = -EINVAL;
+		btrfs_print_v0_err(fs_info);
+		btrfs_abort_transaction(trans, err);
+		goto out;
 	}
-#endif
-	BUG_ON(item_size < sizeof(*ei));
+
 	ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 	__run_delayed_extent_op(extent_op, leaf, ei);
 
@@ -2422,7 +2303,6 @@ out:
 }
 
 static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
 				struct btrfs_delayed_ref_node *node,
 				struct btrfs_delayed_extent_op *extent_op,
 				int insert_reserved)
@@ -2433,14 +2313,14 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 	u64 ref_root = 0;
 
 	ref = btrfs_delayed_node_to_tree_ref(node);
-	trace_run_delayed_tree_ref(fs_info, node, ref, node->action);
+	trace_run_delayed_tree_ref(trans->fs_info, node, ref, node->action);
 
 	if (node->type == BTRFS_SHARED_BLOCK_REF_KEY)
 		parent = ref->parent;
 	ref_root = ref->root;
 
 	if (node->ref_mod != 1) {
-		btrfs_err(fs_info,
+		btrfs_err(trans->fs_info,
 	"btree block(%llu) has %d references rather than 1: action %d ref_root %llu parent %llu",
 			  node->bytenr, node->ref_mod, node->action, ref_root,
 			  parent);
@@ -2450,13 +2330,10 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 		BUG_ON(!extent_op || !extent_op->update_flags);
 		ret = alloc_reserved_tree_block(trans, node, extent_op);
 	} else if (node->action == BTRFS_ADD_DELAYED_REF) {
-		ret = __btrfs_inc_extent_ref(trans, fs_info, node,
-					     parent, ref_root,
-					     ref->level, 0, 1,
-					     extent_op);
+		ret = __btrfs_inc_extent_ref(trans, node, parent, ref_root,
+					     ref->level, 0, 1, extent_op);
 	} else if (node->action == BTRFS_DROP_DELAYED_REF) {
-		ret = __btrfs_free_extent(trans, fs_info, node,
-					  parent, ref_root,
+		ret = __btrfs_free_extent(trans, node, parent, ref_root,
 					  ref->level, 0, 1, extent_op);
 	} else {
 		BUG();
@@ -2466,7 +2343,6 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 
 /* helper function to actually process a single delayed ref entry */
 static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
-			       struct btrfs_fs_info *fs_info,
 			       struct btrfs_delayed_ref_node *node,
 			       struct btrfs_delayed_extent_op *extent_op,
 			       int insert_reserved)
@@ -2475,18 +2351,18 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
 
 	if (trans->aborted) {
 		if (insert_reserved)
-			btrfs_pin_extent(fs_info, node->bytenr,
+			btrfs_pin_extent(trans->fs_info, node->bytenr,
 					 node->num_bytes, 1);
 		return 0;
 	}
 
 	if (node->type == BTRFS_TREE_BLOCK_REF_KEY ||
 	    node->type == BTRFS_SHARED_BLOCK_REF_KEY)
-		ret = run_delayed_tree_ref(trans, fs_info, node, extent_op,
+		ret = run_delayed_tree_ref(trans, node, extent_op,
 					   insert_reserved);
 	else if (node->type == BTRFS_EXTENT_DATA_REF_KEY ||
 		 node->type == BTRFS_SHARED_DATA_REF_KEY)
-		ret = run_delayed_data_ref(trans, fs_info, node, extent_op,
+		ret = run_delayed_data_ref(trans, node, extent_op,
 					   insert_reserved);
 	else
 		BUG();
@@ -2528,7 +2404,6 @@ static void unselect_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_ref
 }
 
 static int cleanup_extent_op(struct btrfs_trans_handle *trans,
-			     struct btrfs_fs_info *fs_info,
 			     struct btrfs_delayed_ref_head *head)
 {
 	struct btrfs_delayed_extent_op *extent_op = head->extent_op;
@@ -2542,21 +2417,22 @@ static int cleanup_extent_op(struct btrfs_trans_handle *trans,
 		return 0;
 	}
 	spin_unlock(&head->lock);
-	ret = run_delayed_extent_op(trans, fs_info, head, extent_op);
+	ret = run_delayed_extent_op(trans, head, extent_op);
 	btrfs_free_delayed_extent_op(extent_op);
 	return ret ? ret : 1;
 }
 
 static int cleanup_ref_head(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info,
 			    struct btrfs_delayed_ref_head *head)
 {
+
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	int ret;
 
 	delayed_refs = &trans->transaction->delayed_refs;
 
-	ret = cleanup_extent_op(trans, fs_info, head);
+	ret = cleanup_extent_op(trans, head);
 	if (ret < 0) {
 		unselect_delayed_ref_head(delayed_refs, head);
 		btrfs_debug(fs_info, "run_delayed_extent_op returned %d", ret);
@@ -2598,8 +2474,9 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans,
 			flags = BTRFS_BLOCK_GROUP_METADATA;
 		space_info = __find_space_info(fs_info, flags);
 		ASSERT(space_info);
-		percpu_counter_add(&space_info->total_bytes_pinned,
-				   -head->num_bytes);
+		percpu_counter_add_batch(&space_info->total_bytes_pinned,
+				   -head->num_bytes,
+				   BTRFS_TOTAL_BYTES_PINNED_BATCH);
 
 		if (head->is_data) {
 			spin_lock(&delayed_refs->lock);
@@ -2705,7 +2582,7 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 		 * up and move on to the next ref_head.
 		 */
 		if (!ref) {
-			ret = cleanup_ref_head(trans, fs_info, locked_ref);
+			ret = cleanup_ref_head(trans, locked_ref);
 			if (ret > 0 ) {
 				/* We dropped our lock, we need to loop. */
 				ret = 0;
@@ -2752,7 +2629,7 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 		locked_ref->extent_op = NULL;
 		spin_unlock(&locked_ref->lock);
 
-		ret = run_one_delayed_ref(trans, fs_info, ref, extent_op,
+		ret = run_one_delayed_ref(trans, ref, extent_op,
 					  must_insert_reserved);
 
 		btrfs_free_delayed_extent_op(extent_op);
@@ -3227,12 +3104,6 @@ static noinline int check_committed_ref(struct btrfs_root *root,
 
 	ret = 1;
 	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (item_size < sizeof(*ei)) {
-		WARN_ON(item_size != sizeof(struct btrfs_extent_item_v0));
-		goto out;
-	}
-#endif
 	ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 
 	if (item_size != sizeof(*ei) +
@@ -4060,11 +3931,7 @@ static void update_space_info(struct btrfs_fs_info *info, u64 flags,
 	struct btrfs_space_info *found;
 	int factor;
 
-	if (flags & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1 |
-		     BTRFS_BLOCK_GROUP_RAID10))
-		factor = 2;
-	else
-		factor = 1;
+	factor = btrfs_bg_type_to_factor(flags);
 
 	found = __find_space_info(info, flags);
 	ASSERT(found);
@@ -4289,7 +4156,7 @@ again:
 			if (IS_ERR(trans))
 				return PTR_ERR(trans);
 
-			ret = do_chunk_alloc(trans, fs_info, alloc_target,
+			ret = do_chunk_alloc(trans, alloc_target,
 					     CHUNK_ALLOC_NO_FORCE);
 			btrfs_end_transaction(trans);
 			if (ret < 0) {
@@ -4309,9 +4176,10 @@ again:
 		 * allocation, and no removed chunk in current transaction,
 		 * don't bother committing the transaction.
 		 */
-		have_pinned_space = percpu_counter_compare(
+		have_pinned_space = __percpu_counter_compare(
 			&data_sinfo->total_bytes_pinned,
-			used + bytes - data_sinfo->total_bytes);
+			used + bytes - data_sinfo->total_bytes,
+			BTRFS_TOTAL_BYTES_PINNED_BATCH);
 		spin_unlock(&data_sinfo->lock);
 
 		/* commit the current transaction and try again */
@@ -4358,7 +4226,7 @@ commit_trans:
 				      data_sinfo->flags, bytes, 1);
 	spin_unlock(&data_sinfo->lock);
 
-	return ret;
+	return 0;
 }
 
 int btrfs_check_data_free_space(struct inode *inode,
@@ -4511,9 +4379,9 @@ static u64 get_profile_num_devs(struct btrfs_fs_info *fs_info, u64 type)
  * for allocating a chunk, otherwise if it's false, reserve space necessary for
  * removing a chunk.
  */
-void check_system_chunk(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u64 type)
+void check_system_chunk(struct btrfs_trans_handle *trans, u64 type)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_space_info *info;
 	u64 left;
 	u64 thresh;
@@ -4552,7 +4420,7 @@ void check_system_chunk(struct btrfs_trans_handle *trans,
 		 * the paths we visit in the chunk tree (they were already COWed
 		 * or created in the current transaction for example).
 		 */
-		ret = btrfs_alloc_chunk(trans, fs_info, flags);
+		ret = btrfs_alloc_chunk(trans, flags);
 	}
 
 	if (!ret) {
@@ -4573,11 +4441,13 @@ void check_system_chunk(struct btrfs_trans_handle *trans,
  *    - return 1 if it successfully allocates a chunk,
  *    - return errors including -ENOSPC otherwise.
  */
-static int do_chunk_alloc(struct btrfs_trans_handle *trans,
-			  struct btrfs_fs_info *fs_info, u64 flags, int force)
+static int do_chunk_alloc(struct btrfs_trans_handle *trans, u64 flags,
+			  int force)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_space_info *space_info;
-	int wait_for_alloc = 0;
+	bool wait_for_alloc = false;
+	bool should_alloc = false;
 	int ret = 0;
 
 	/* Don't re-enter if we're already allocating a chunk */
@@ -4587,45 +4457,44 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 	space_info = __find_space_info(fs_info, flags);
 	ASSERT(space_info);
 
-again:
-	spin_lock(&space_info->lock);
-	if (force < space_info->force_alloc)
-		force = space_info->force_alloc;
-	if (space_info->full) {
-		if (should_alloc_chunk(fs_info, space_info, force))
-			ret = -ENOSPC;
-		else
-			ret = 0;
-		spin_unlock(&space_info->lock);
-		return ret;
-	}
-
-	if (!should_alloc_chunk(fs_info, space_info, force)) {
-		spin_unlock(&space_info->lock);
-		return 0;
-	} else if (space_info->chunk_alloc) {
-		wait_for_alloc = 1;
-	} else {
-		space_info->chunk_alloc = 1;
-	}
-
-	spin_unlock(&space_info->lock);
-
-	mutex_lock(&fs_info->chunk_mutex);
+	do {
+		spin_lock(&space_info->lock);
+		if (force < space_info->force_alloc)
+			force = space_info->force_alloc;
+		should_alloc = should_alloc_chunk(fs_info, space_info, force);
+		if (space_info->full) {
+			/* No more free physical space */
+			if (should_alloc)
+				ret = -ENOSPC;
+			else
+				ret = 0;
+			spin_unlock(&space_info->lock);
+			return ret;
+		} else if (!should_alloc) {
+			spin_unlock(&space_info->lock);
+			return 0;
+		} else if (space_info->chunk_alloc) {
+			/*
+			 * Someone is already allocating, so we need to block
+			 * until this someone is finished and then loop to
+			 * recheck if we should continue with our allocation
+			 * attempt.
+			 */
+			wait_for_alloc = true;
+			spin_unlock(&space_info->lock);
+			mutex_lock(&fs_info->chunk_mutex);
+			mutex_unlock(&fs_info->chunk_mutex);
+		} else {
+			/* Proceed with allocation */
+			space_info->chunk_alloc = 1;
+			wait_for_alloc = false;
+			spin_unlock(&space_info->lock);
+		}
 
-	/*
-	 * The chunk_mutex is held throughout the entirety of a chunk
-	 * allocation, so once we've acquired the chunk_mutex we know that the
-	 * other guy is done and we need to recheck and see if we should
-	 * allocate.
-	 */
-	if (wait_for_alloc) {
-		mutex_unlock(&fs_info->chunk_mutex);
-		wait_for_alloc = 0;
 		cond_resched();
-		goto again;
-	}
+	} while (wait_for_alloc);
 
+	mutex_lock(&fs_info->chunk_mutex);
 	trans->allocating_chunk = true;
 
 	/*
@@ -4651,9 +4520,9 @@ again:
 	 * Check if we have enough space in SYSTEM chunk because we may need
 	 * to update devices.
 	 */
-	check_system_chunk(trans, fs_info, flags);
+	check_system_chunk(trans, flags);
 
-	ret = btrfs_alloc_chunk(trans, fs_info, flags);
+	ret = btrfs_alloc_chunk(trans, flags);
 	trans->allocating_chunk = false;
 
 	spin_lock(&space_info->lock);
@@ -4703,6 +4572,7 @@ static int can_overcommit(struct btrfs_fs_info *fs_info,
 	u64 space_size;
 	u64 avail;
 	u64 used;
+	int factor;
 
 	/* Don't overcommit when in mixed mode. */
 	if (space_info->flags & BTRFS_BLOCK_GROUP_DATA)
@@ -4737,10 +4607,8 @@ static int can_overcommit(struct btrfs_fs_info *fs_info,
 	 * doesn't include the parity drive, so we don't have to
 	 * change the math
 	 */
-	if (profile & (BTRFS_BLOCK_GROUP_DUP |
-		       BTRFS_BLOCK_GROUP_RAID1 |
-		       BTRFS_BLOCK_GROUP_RAID10))
-		avail >>= 1;
+	factor = btrfs_bg_type_to_factor(profile);
+	avail = div_u64(avail, factor);
 
 	/*
 	 * If we aren't flushing all things, let us overcommit up to
@@ -4912,8 +4780,9 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 		return 0;
 
 	/* See if there is enough pinned space to make this reservation */
-	if (percpu_counter_compare(&space_info->total_bytes_pinned,
-				   bytes) >= 0)
+	if (__percpu_counter_compare(&space_info->total_bytes_pinned,
+				   bytes,
+				   BTRFS_TOTAL_BYTES_PINNED_BATCH) >= 0)
 		goto commit;
 
 	/*
@@ -4930,8 +4799,9 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 		bytes -= delayed_rsv->size;
 	spin_unlock(&delayed_rsv->lock);
 
-	if (percpu_counter_compare(&space_info->total_bytes_pinned,
-				   bytes) < 0) {
+	if (__percpu_counter_compare(&space_info->total_bytes_pinned,
+				   bytes,
+				   BTRFS_TOTAL_BYTES_PINNED_BATCH) < 0) {
 		return -ENOSPC;
 	}
 
@@ -4984,7 +4854,7 @@ static void flush_space(struct btrfs_fs_info *fs_info,
 			ret = PTR_ERR(trans);
 			break;
 		}
-		ret = do_chunk_alloc(trans, fs_info,
+		ret = do_chunk_alloc(trans,
 				     btrfs_metadata_alloc_profile(fs_info),
 				     CHUNK_ALLOC_NO_FORCE);
 		btrfs_end_transaction(trans);
@@ -5659,11 +5529,6 @@ void btrfs_free_block_rsv(struct btrfs_fs_info *fs_info,
 	kfree(rsv);
 }
 
-void __btrfs_free_block_rsv(struct btrfs_block_rsv *rsv)
-{
-	kfree(rsv);
-}
-
 int btrfs_block_rsv_add(struct btrfs_root *root,
 			struct btrfs_block_rsv *block_rsv, u64 num_bytes,
 			enum btrfs_reserve_flush_enum flush)
@@ -6019,7 +5884,7 @@ static void btrfs_calculate_inode_block_rsv_size(struct btrfs_fs_info *fs_info,
 
 int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	unsigned nr_extents;
 	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL;
 	int ret = 0;
@@ -6092,7 +5957,7 @@ out_fail:
 void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes,
 				     bool qgroup_free)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 
 	num_bytes = ALIGN(num_bytes, fs_info->sectorsize);
 	spin_lock(&inode->lock);
@@ -6121,7 +5986,7 @@ void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes,
 void btrfs_delalloc_release_extents(struct btrfs_inode *inode, u64 num_bytes,
 				    bool qgroup_free)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	unsigned num_extents;
 
 	spin_lock(&inode->lock);
@@ -6219,12 +6084,8 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 		cache = btrfs_lookup_block_group(info, bytenr);
 		if (!cache)
 			return -ENOENT;
-		if (cache->flags & (BTRFS_BLOCK_GROUP_DUP |
-				    BTRFS_BLOCK_GROUP_RAID1 |
-				    BTRFS_BLOCK_GROUP_RAID10))
-			factor = 2;
-		else
-			factor = 1;
+		factor = btrfs_bg_type_to_factor(cache->flags);
+
 		/*
 		 * If this block group has free space cache written out, we
 		 * need to make sure to load it if we are removing space.  This
@@ -6268,8 +6129,9 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 			trace_btrfs_space_reservation(info, "pinned",
 						      cache->space_info->flags,
 						      num_bytes, 1);
-			percpu_counter_add(&cache->space_info->total_bytes_pinned,
-					   num_bytes);
+			percpu_counter_add_batch(&cache->space_info->total_bytes_pinned,
+					   num_bytes,
+					   BTRFS_TOTAL_BYTES_PINNED_BATCH);
 			set_extent_dirty(info->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
@@ -6279,7 +6141,7 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 		if (list_empty(&cache->dirty_list)) {
 			list_add_tail(&cache->dirty_list,
 				      &trans->transaction->dirty_bgs);
-				trans->transaction->num_dirty_bgs++;
+			trans->transaction->num_dirty_bgs++;
 			btrfs_get_block_group(cache);
 		}
 		spin_unlock(&trans->transaction->dirty_bgs_lock);
@@ -6290,16 +6152,8 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 		 * dirty list to avoid races between cleaner kthread and space
 		 * cache writeout.
 		 */
-		if (!alloc && old_val == 0) {
-			spin_lock(&info->unused_bgs_lock);
-			if (list_empty(&cache->bg_list)) {
-				btrfs_get_block_group(cache);
-				trace_btrfs_add_unused_block_group(cache);
-				list_add_tail(&cache->bg_list,
-					      &info->unused_bgs);
-			}
-			spin_unlock(&info->unused_bgs_lock);
-		}
+		if (!alloc && old_val == 0)
+			btrfs_mark_bg_unused(cache);
 
 		btrfs_put_block_group(cache);
 		total -= num_bytes;
@@ -6347,7 +6201,8 @@ static int pin_down_extent(struct btrfs_fs_info *fs_info,
 
 	trace_btrfs_space_reservation(fs_info, "pinned",
 				      cache->space_info->flags, num_bytes, 1);
-	percpu_counter_add(&cache->space_info->total_bytes_pinned, num_bytes);
+	percpu_counter_add_batch(&cache->space_info->total_bytes_pinned,
+		    num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH);
 	set_extent_dirty(fs_info->pinned_extents, bytenr,
 			 bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
 	return 0;
@@ -6711,7 +6566,8 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info,
 		trace_btrfs_space_reservation(fs_info, "pinned",
 					      space_info->flags, len, 0);
 		space_info->max_extent_size = 0;
-		percpu_counter_add(&space_info->total_bytes_pinned, -len);
+		percpu_counter_add_batch(&space_info->total_bytes_pinned,
+			    -len, BTRFS_TOTAL_BYTES_PINNED_BATCH);
 		if (cache->ro) {
 			space_info->bytes_readonly += len;
 			readonly = true;
@@ -6815,12 +6671,12 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 }
 
 static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *info,
-				struct btrfs_delayed_ref_node *node, u64 parent,
-				u64 root_objectid, u64 owner_objectid,
-				u64 owner_offset, int refs_to_drop,
-				struct btrfs_delayed_extent_op *extent_op)
+			       struct btrfs_delayed_ref_node *node, u64 parent,
+			       u64 root_objectid, u64 owner_objectid,
+			       u64 owner_offset, int refs_to_drop,
+			       struct btrfs_delayed_extent_op *extent_op)
 {
+	struct btrfs_fs_info *info = trans->fs_info;
 	struct btrfs_key key;
 	struct btrfs_path *path;
 	struct btrfs_root *extent_root = info->extent_root;
@@ -6852,9 +6708,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 	if (is_data)
 		skinny_metadata = false;
 
-	ret = lookup_extent_backref(trans, info, path, &iref,
-				    bytenr, num_bytes, parent,
-				    root_objectid, owner_objectid,
+	ret = lookup_extent_backref(trans, path, &iref, bytenr, num_bytes,
+				    parent, root_objectid, owner_objectid,
 				    owner_offset);
 	if (ret == 0) {
 		extent_slot = path->slots[0];
@@ -6877,14 +6732,10 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 				break;
 			extent_slot--;
 		}
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		item_size = btrfs_item_size_nr(path->nodes[0], extent_slot);
-		if (found_extent && item_size < sizeof(*ei))
-			found_extent = 0;
-#endif
+
 		if (!found_extent) {
 			BUG_ON(iref);
-			ret = remove_extent_backref(trans, info, path, NULL,
+			ret = remove_extent_backref(trans, path, NULL,
 						    refs_to_drop,
 						    is_data, &last_ref);
 			if (ret) {
@@ -6957,42 +6808,12 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 
 	leaf = path->nodes[0];
 	item_size = btrfs_item_size_nr(leaf, extent_slot);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (item_size < sizeof(*ei)) {
-		BUG_ON(found_extent || extent_slot != path->slots[0]);
-		ret = convert_extent_item_v0(trans, info, path, owner_objectid,
-					     0);
-		if (ret < 0) {
-			btrfs_abort_transaction(trans, ret);
-			goto out;
-		}
-
-		btrfs_release_path(path);
-		path->leave_spinning = 1;
-
-		key.objectid = bytenr;
-		key.type = BTRFS_EXTENT_ITEM_KEY;
-		key.offset = num_bytes;
-
-		ret = btrfs_search_slot(trans, extent_root, &key, path,
-					-1, 1);
-		if (ret) {
-			btrfs_err(info,
-				  "umm, got %d back from search, was looking for %llu",
-				ret, bytenr);
-			btrfs_print_leaf(path->nodes[0]);
-		}
-		if (ret < 0) {
-			btrfs_abort_transaction(trans, ret);
-			goto out;
-		}
-
-		extent_slot = path->slots[0];
-		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, extent_slot);
+	if (unlikely(item_size < sizeof(*ei))) {
+		ret = -EINVAL;
+		btrfs_print_v0_err(info);
+		btrfs_abort_transaction(trans, ret);
+		goto out;
 	}
-#endif
-	BUG_ON(item_size < sizeof(*ei));
 	ei = btrfs_item_ptr(leaf, extent_slot,
 			    struct btrfs_extent_item);
 	if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID &&
@@ -7028,9 +6849,9 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			btrfs_mark_buffer_dirty(leaf);
 		}
 		if (found_extent) {
-			ret = remove_extent_backref(trans, info, path,
-						    iref, refs_to_drop,
-						    is_data, &last_ref);
+			ret = remove_extent_backref(trans, path, iref,
+						    refs_to_drop, is_data,
+						    &last_ref);
 			if (ret) {
 				btrfs_abort_transaction(trans, ret);
 				goto out;
@@ -7172,7 +6993,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 				   root->root_key.objectid,
 				   btrfs_header_level(buf), 0,
 				   BTRFS_DROP_DELAYED_REF);
-		ret = btrfs_add_delayed_tree_ref(fs_info, trans, buf->start,
+		ret = btrfs_add_delayed_tree_ref(trans, buf->start,
 						 buf->len, parent,
 						 root->root_key.objectid,
 						 btrfs_header_level(buf),
@@ -7251,13 +7072,13 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans,
 		old_ref_mod = new_ref_mod = 0;
 		ret = 0;
 	} else if (owner < BTRFS_FIRST_FREE_OBJECTID) {
-		ret = btrfs_add_delayed_tree_ref(fs_info, trans, bytenr,
+		ret = btrfs_add_delayed_tree_ref(trans, bytenr,
 						 num_bytes, parent,
 						 root_objectid, (int)owner,
 						 BTRFS_DROP_DELAYED_REF, NULL,
 						 &old_ref_mod, &new_ref_mod);
 	} else {
-		ret = btrfs_add_delayed_data_ref(fs_info, trans, bytenr,
+		ret = btrfs_add_delayed_data_ref(trans, bytenr,
 						 num_bytes, parent,
 						 root_objectid, owner, offset,
 						 0, BTRFS_DROP_DELAYED_REF,
@@ -7534,7 +7355,7 @@ search:
 		 * for the proper type.
 		 */
 		if (!block_group_bits(block_group, flags)) {
-		    u64 extra = BTRFS_BLOCK_GROUP_DUP |
+			u64 extra = BTRFS_BLOCK_GROUP_DUP |
 				BTRFS_BLOCK_GROUP_RAID1 |
 				BTRFS_BLOCK_GROUP_RAID5 |
 				BTRFS_BLOCK_GROUP_RAID6 |
@@ -7738,7 +7559,7 @@ unclustered_alloc:
 			goto loop;
 		}
 checks:
-		search_start = ALIGN(offset, fs_info->stripesize);
+		search_start = round_up(offset, fs_info->stripesize);
 
 		/* move on to the next group */
 		if (search_start + num_bytes >
@@ -7750,7 +7571,6 @@ checks:
 		if (offset < search_start)
 			btrfs_add_free_space(block_group, offset,
 					     search_start - offset);
-		BUG_ON(offset > search_start);
 
 		ret = btrfs_add_reserved_bytes(block_group, ram_bytes,
 				num_bytes, delalloc);
@@ -7826,8 +7646,7 @@ loop:
 				goto out;
 			}
 
-			ret = do_chunk_alloc(trans, fs_info, flags,
-					     CHUNK_ALLOC_FORCE);
+			ret = do_chunk_alloc(trans, flags, CHUNK_ALLOC_FORCE);
 
 			/*
 			 * If we can't allocate a new chunk we've already looped
@@ -8053,11 +7872,11 @@ int btrfs_free_and_pin_reserved_extent(struct btrfs_fs_info *fs_info,
 }
 
 static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
-				      struct btrfs_fs_info *fs_info,
 				      u64 parent, u64 root_objectid,
 				      u64 flags, u64 owner, u64 offset,
 				      struct btrfs_key *ins, int ref_mod)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
 	struct btrfs_extent_item *extent_item;
 	struct btrfs_extent_inline_ref *iref;
@@ -8231,7 +8050,6 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 				     u64 offset, u64 ram_bytes,
 				     struct btrfs_key *ins)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
 
 	BUG_ON(root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID);
@@ -8240,7 +8058,7 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 			   root->root_key.objectid, owner, offset,
 			   BTRFS_ADD_DELAYED_EXTENT);
 
-	ret = btrfs_add_delayed_data_ref(fs_info, trans, ins->objectid,
+	ret = btrfs_add_delayed_data_ref(trans, ins->objectid,
 					 ins->offset, 0,
 					 root->root_key.objectid, owner,
 					 offset, ram_bytes,
@@ -8254,10 +8072,10 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
  * space cache bits as well
  */
 int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   u64 root_objectid, u64 owner, u64 offset,
 				   struct btrfs_key *ins)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
 	struct btrfs_block_group_cache *block_group;
 	struct btrfs_space_info *space_info;
@@ -8285,15 +8103,15 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 	spin_unlock(&block_group->lock);
 	spin_unlock(&space_info->lock);
 
-	ret = alloc_reserved_file_extent(trans, fs_info, 0, root_objectid,
-					 0, owner, offset, ins, 1);
+	ret = alloc_reserved_file_extent(trans, 0, root_objectid, 0, owner,
+					 offset, ins, 1);
 	btrfs_put_block_group(block_group);
 	return ret;
 }
 
 static struct extent_buffer *
 btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		      u64 bytenr, int level)
+		      u64 bytenr, int level, u64 owner)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *buf;
@@ -8302,7 +8120,6 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	if (IS_ERR(buf))
 		return buf;
 
-	btrfs_set_header_generation(buf, trans->transid);
 	btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
 	btrfs_tree_lock(buf);
 	clean_tree_block(fs_info, buf);
@@ -8311,6 +8128,14 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	btrfs_set_lock_blocking(buf);
 	set_extent_buffer_uptodate(buf);
 
+	memzero_extent_buffer(buf, 0, sizeof(struct btrfs_header));
+	btrfs_set_header_level(buf, level);
+	btrfs_set_header_bytenr(buf, buf->start);
+	btrfs_set_header_generation(buf, trans->transid);
+	btrfs_set_header_backref_rev(buf, BTRFS_MIXED_BACKREF_REV);
+	btrfs_set_header_owner(buf, owner);
+	write_extent_buffer_fsid(buf, fs_info->fsid);
+	write_extent_buffer_chunk_tree_uuid(buf, fs_info->chunk_tree_uuid);
 	if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
 		buf->log_index = root->log_transid % 2;
 		/*
@@ -8419,7 +8244,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	if (btrfs_is_testing(fs_info)) {
 		buf = btrfs_init_new_buffer(trans, root, root->alloc_bytenr,
-					    level);
+					    level, root_objectid);
 		if (!IS_ERR(buf))
 			root->alloc_bytenr += blocksize;
 		return buf;
@@ -8435,7 +8260,8 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto out_unuse;
 
-	buf = btrfs_init_new_buffer(trans, root, ins.objectid, level);
+	buf = btrfs_init_new_buffer(trans, root, ins.objectid, level,
+				    root_objectid);
 	if (IS_ERR(buf)) {
 		ret = PTR_ERR(buf);
 		goto out_free_reserved;
@@ -8467,7 +8293,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 		btrfs_ref_tree_mod(root, ins.objectid, ins.offset, parent,
 				   root_objectid, level, 0,
 				   BTRFS_ADD_DELAYED_EXTENT);
-		ret = btrfs_add_delayed_tree_ref(fs_info, trans, ins.objectid,
+		ret = btrfs_add_delayed_tree_ref(trans, ins.objectid,
 						 ins.offset, parent,
 						 root_objectid, level,
 						 BTRFS_ADD_DELAYED_EXTENT,
@@ -8499,7 +8325,6 @@ struct walk_control {
 	int keep_locks;
 	int reada_slot;
 	int reada_count;
-	int for_reloc;
 };
 
 #define DROP_REFERENCE	1
@@ -8819,7 +8644,7 @@ skip:
 		}
 
 		if (need_account) {
-			ret = btrfs_qgroup_trace_subtree(trans, root, next,
+			ret = btrfs_qgroup_trace_subtree(trans, next,
 							 generation, level - 1);
 			if (ret) {
 				btrfs_err_rl(fs_info,
@@ -8919,7 +8744,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
 			else
 				ret = btrfs_dec_ref(trans, root, eb, 0);
 			BUG_ON(ret); /* -ENOMEM */
-			ret = btrfs_qgroup_trace_leaf_items(trans, fs_info, eb);
+			ret = btrfs_qgroup_trace_leaf_items(trans, eb);
 			if (ret) {
 				btrfs_err_rl(fs_info,
 					     "error %d accounting leaf items. Quota is out of sync, rescan required.",
@@ -9136,7 +8961,6 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	wc->stage = DROP_REFERENCE;
 	wc->update_ref = update_ref;
 	wc->keep_locks = 0;
-	wc->for_reloc = for_reloc;
 	wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(fs_info);
 
 	while (1) {
@@ -9199,7 +9023,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	if (err)
 		goto out_end_trans;
 
-	ret = btrfs_del_root(trans, fs_info, &root->root_key);
+	ret = btrfs_del_root(trans, &root->root_key);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		err = ret;
@@ -9302,7 +9126,6 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 	wc->stage = DROP_REFERENCE;
 	wc->update_ref = 0;
 	wc->keep_locks = 1;
-	wc->for_reloc = 1;
 	wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(fs_info);
 
 	while (1) {
@@ -9417,10 +9240,10 @@ out:
 	return ret;
 }
 
-int btrfs_inc_block_group_ro(struct btrfs_fs_info *fs_info,
-			     struct btrfs_block_group_cache *cache)
+int btrfs_inc_block_group_ro(struct btrfs_block_group_cache *cache)
 
 {
+	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct btrfs_trans_handle *trans;
 	u64 alloc_flags;
 	int ret;
@@ -9454,7 +9277,7 @@ again:
 	 */
 	alloc_flags = update_block_group_flags(fs_info, cache->flags);
 	if (alloc_flags != cache->flags) {
-		ret = do_chunk_alloc(trans, fs_info, alloc_flags,
+		ret = do_chunk_alloc(trans, alloc_flags,
 				     CHUNK_ALLOC_FORCE);
 		/*
 		 * ENOSPC is allowed here, we may have enough space
@@ -9471,8 +9294,7 @@ again:
 	if (!ret)
 		goto out;
 	alloc_flags = get_alloc_profile(fs_info, cache->space_info->flags);
-	ret = do_chunk_alloc(trans, fs_info, alloc_flags,
-			     CHUNK_ALLOC_FORCE);
+	ret = do_chunk_alloc(trans, alloc_flags, CHUNK_ALLOC_FORCE);
 	if (ret < 0)
 		goto out;
 	ret = inc_block_group_ro(cache, 0);
@@ -9480,7 +9302,7 @@ out:
 	if (cache->flags & BTRFS_BLOCK_GROUP_SYSTEM) {
 		alloc_flags = update_block_group_flags(fs_info, cache->flags);
 		mutex_lock(&fs_info->chunk_mutex);
-		check_system_chunk(trans, fs_info, alloc_flags);
+		check_system_chunk(trans, alloc_flags);
 		mutex_unlock(&fs_info->chunk_mutex);
 	}
 	mutex_unlock(&fs_info->ro_block_group_mutex);
@@ -9489,12 +9311,11 @@ out:
 	return ret;
 }
 
-int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info, u64 type)
+int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans, u64 type)
 {
-	u64 alloc_flags = get_alloc_profile(fs_info, type);
+	u64 alloc_flags = get_alloc_profile(trans->fs_info, type);
 
-	return do_chunk_alloc(trans, fs_info, alloc_flags, CHUNK_ALLOC_FORCE);
+	return do_chunk_alloc(trans, alloc_flags, CHUNK_ALLOC_FORCE);
 }
 
 /*
@@ -9520,13 +9341,7 @@ u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo)
 			continue;
 		}
 
-		if (block_group->flags & (BTRFS_BLOCK_GROUP_RAID1 |
-					  BTRFS_BLOCK_GROUP_RAID10 |
-					  BTRFS_BLOCK_GROUP_DUP))
-			factor = 2;
-		else
-			factor = 1;
-
+		factor = btrfs_bg_type_to_factor(block_group->flags);
 		free_bytes += (block_group->key.offset -
 			       btrfs_block_group_used(&block_group->item)) *
 			       factor;
@@ -9717,6 +9532,8 @@ static int find_first_block_group(struct btrfs_fs_info *fs_info,
 	int ret = 0;
 	struct btrfs_key found_key;
 	struct extent_buffer *leaf;
+	struct btrfs_block_group_item bg;
+	u64 flags;
 	int slot;
 
 	ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
@@ -9751,8 +9568,32 @@ static int find_first_block_group(struct btrfs_fs_info *fs_info,
 			"logical %llu len %llu found bg but no related chunk",
 					  found_key.objectid, found_key.offset);
 				ret = -ENOENT;
+			} else if (em->start != found_key.objectid ||
+				   em->len != found_key.offset) {
+				btrfs_err(fs_info,
+		"block group %llu len %llu mismatch with chunk %llu len %llu",
+					  found_key.objectid, found_key.offset,
+					  em->start, em->len);
+				ret = -EUCLEAN;
 			} else {
-				ret = 0;
+				read_extent_buffer(leaf, &bg,
+					btrfs_item_ptr_offset(leaf, slot),
+					sizeof(bg));
+				flags = btrfs_block_group_flags(&bg) &
+					BTRFS_BLOCK_GROUP_TYPE_MASK;
+
+				if (flags != (em->map_lookup->type &
+					      BTRFS_BLOCK_GROUP_TYPE_MASK)) {
+					btrfs_err(fs_info,
+"block group %llu len %llu type flags 0x%llx mismatch with chunk type flags 0x%llx",
+						found_key.objectid,
+						found_key.offset, flags,
+						(BTRFS_BLOCK_GROUP_TYPE_MASK &
+						 em->map_lookup->type));
+					ret = -EUCLEAN;
+				} else {
+					ret = 0;
+				}
 			}
 			free_extent_map(em);
 			goto out;
@@ -9847,7 +9688,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		 */
 		if (block_group->cached == BTRFS_CACHE_NO ||
 		    block_group->cached == BTRFS_CACHE_ERROR)
-			free_excluded_extents(info, block_group);
+			free_excluded_extents(block_group);
 
 		btrfs_remove_free_space_cache(block_group);
 		ASSERT(block_group->cached != BTRFS_CACHE_STARTED);
@@ -10003,6 +9844,62 @@ btrfs_create_block_group_cache(struct btrfs_fs_info *fs_info,
 	return cache;
 }
 
+
+/*
+ * Iterate all chunks and verify that each of them has the corresponding block
+ * group
+ */
+static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
+	struct extent_map *em;
+	struct btrfs_block_group_cache *bg;
+	u64 start = 0;
+	int ret = 0;
+
+	while (1) {
+		read_lock(&map_tree->map_tree.lock);
+		/*
+		 * lookup_extent_mapping will return the first extent map
+		 * intersecting the range, so setting @len to 1 is enough to
+		 * get the first chunk.
+		 */
+		em = lookup_extent_mapping(&map_tree->map_tree, start, 1);
+		read_unlock(&map_tree->map_tree.lock);
+		if (!em)
+			break;
+
+		bg = btrfs_lookup_block_group(fs_info, em->start);
+		if (!bg) {
+			btrfs_err(fs_info,
+	"chunk start=%llu len=%llu doesn't have corresponding block group",
+				     em->start, em->len);
+			ret = -EUCLEAN;
+			free_extent_map(em);
+			break;
+		}
+		if (bg->key.objectid != em->start ||
+		    bg->key.offset != em->len ||
+		    (bg->flags & BTRFS_BLOCK_GROUP_TYPE_MASK) !=
+		    (em->map_lookup->type & BTRFS_BLOCK_GROUP_TYPE_MASK)) {
+			btrfs_err(fs_info,
+"chunk start=%llu len=%llu flags=0x%llx doesn't match block group start=%llu len=%llu flags=0x%llx",
+				em->start, em->len,
+				em->map_lookup->type & BTRFS_BLOCK_GROUP_TYPE_MASK,
+				bg->key.objectid, bg->key.offset,
+				bg->flags & BTRFS_BLOCK_GROUP_TYPE_MASK);
+			ret = -EUCLEAN;
+			free_extent_map(em);
+			btrfs_put_block_group(bg);
+			break;
+		}
+		start = em->start + em->len;
+		free_extent_map(em);
+		btrfs_put_block_group(bg);
+	}
+	return ret;
+}
+
 int btrfs_read_block_groups(struct btrfs_fs_info *info)
 {
 	struct btrfs_path *path;
@@ -10089,13 +9986,13 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		 * info has super bytes accounted for, otherwise we'll think
 		 * we have more space than we actually do.
 		 */
-		ret = exclude_super_stripes(info, cache);
+		ret = exclude_super_stripes(cache);
 		if (ret) {
 			/*
 			 * We may have excluded something, so call this just in
 			 * case.
 			 */
-			free_excluded_extents(info, cache);
+			free_excluded_extents(cache);
 			btrfs_put_block_group(cache);
 			goto error;
 		}
@@ -10110,14 +10007,14 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		if (found_key.offset == btrfs_block_group_used(&cache->item)) {
 			cache->last_byte_to_unpin = (u64)-1;
 			cache->cached = BTRFS_CACHE_FINISHED;
-			free_excluded_extents(info, cache);
+			free_excluded_extents(cache);
 		} else if (btrfs_block_group_used(&cache->item) == 0) {
 			cache->last_byte_to_unpin = (u64)-1;
 			cache->cached = BTRFS_CACHE_FINISHED;
 			add_new_free_space(cache, found_key.objectid,
 					   found_key.objectid +
 					   found_key.offset);
-			free_excluded_extents(info, cache);
+			free_excluded_extents(cache);
 		}
 
 		ret = btrfs_add_block_group_cache(info, cache);
@@ -10140,15 +10037,8 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		if (btrfs_chunk_readonly(info, cache->key.objectid)) {
 			inc_block_group_ro(cache, 1);
 		} else if (btrfs_block_group_used(&cache->item) == 0) {
-			spin_lock(&info->unused_bgs_lock);
-			/* Should always be true but just in case. */
-			if (list_empty(&cache->bg_list)) {
-				btrfs_get_block_group(cache);
-				trace_btrfs_add_unused_block_group(cache);
-				list_add_tail(&cache->bg_list,
-					      &info->unused_bgs);
-			}
-			spin_unlock(&info->unused_bgs_lock);
+			ASSERT(list_empty(&cache->bg_list));
+			btrfs_mark_bg_unused(cache);
 		}
 	}
 
@@ -10176,7 +10066,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 
 	btrfs_add_raid_kobjects(info);
 	init_global_block_rsv(info);
-	ret = 0;
+	ret = check_chunk_block_group_mappings(info);
 error:
 	btrfs_free_path(path);
 	return ret;
@@ -10206,8 +10096,7 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 					sizeof(item));
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
-		ret = btrfs_finish_chunk_alloc(trans, fs_info, key.objectid,
-					       key.offset);
+		ret = btrfs_finish_chunk_alloc(trans, key.objectid, key.offset);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
 		add_block_group_free_space(trans, block_group);
@@ -10218,10 +10107,10 @@ next:
 	trans->can_flush_pending_bgs = can_flush_pending_bgs;
 }
 
-int btrfs_make_block_group(struct btrfs_trans_handle *trans,
-			   struct btrfs_fs_info *fs_info, u64 bytes_used,
+int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 			   u64 type, u64 chunk_offset, u64 size)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_block_group_cache *cache;
 	int ret;
 
@@ -10240,20 +10129,20 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	cache->last_byte_to_unpin = (u64)-1;
 	cache->cached = BTRFS_CACHE_FINISHED;
 	cache->needs_free_space = 1;
-	ret = exclude_super_stripes(fs_info, cache);
+	ret = exclude_super_stripes(cache);
 	if (ret) {
 		/*
 		 * We may have excluded something, so call this just in
 		 * case.
 		 */
-		free_excluded_extents(fs_info, cache);
+		free_excluded_extents(cache);
 		btrfs_put_block_group(cache);
 		return ret;
 	}
 
 	add_new_free_space(cache, chunk_offset, chunk_offset + size);
 
-	free_excluded_extents(fs_info, cache);
+	free_excluded_extents(cache);
 
 #ifdef CONFIG_BTRFS_DEBUG
 	if (btrfs_should_fragment_free_space(cache)) {
@@ -10311,9 +10200,9 @@ static void clear_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
 }
 
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
-			     struct btrfs_fs_info *fs_info, u64 group_start,
-			     struct extent_map *em)
+			     u64 group_start, struct extent_map *em)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = fs_info->extent_root;
 	struct btrfs_path *path;
 	struct btrfs_block_group_cache *block_group;
@@ -10337,18 +10226,13 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	 * Free the reserved super bytes from this block group before
 	 * remove it.
 	 */
-	free_excluded_extents(fs_info, block_group);
+	free_excluded_extents(block_group);
 	btrfs_free_ref_tree_range(fs_info, block_group->key.objectid,
 				  block_group->key.offset);
 
 	memcpy(&key, &block_group->key, sizeof(key));
 	index = btrfs_bg_flags_to_raid_index(block_group->flags);
-	if (block_group->flags & (BTRFS_BLOCK_GROUP_DUP |
-				  BTRFS_BLOCK_GROUP_RAID1 |
-				  BTRFS_BLOCK_GROUP_RAID10))
-		factor = 2;
-	else
-		factor = 1;
+	factor = btrfs_bg_type_to_factor(block_group->flags);
 
 	/* make sure this block group isn't part of an allocation cluster */
 	cluster = &fs_info->data_alloc_cluster;
@@ -10687,7 +10571,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		/* Don't want to race with allocators so take the groups_sem */
 		down_write(&space_info->groups_sem);
 		spin_lock(&block_group->lock);
-		if (block_group->reserved ||
+		if (block_group->reserved || block_group->pinned ||
 		    btrfs_block_group_used(&block_group->item) ||
 		    block_group->ro ||
 		    list_is_singular(&block_group->list)) {
@@ -10764,8 +10648,9 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 		space_info->bytes_pinned -= block_group->pinned;
 		space_info->bytes_readonly += block_group->pinned;
-		percpu_counter_add(&space_info->total_bytes_pinned,
-				   -block_group->pinned);
+		percpu_counter_add_batch(&space_info->total_bytes_pinned,
+				   -block_group->pinned,
+				   BTRFS_TOTAL_BYTES_PINNED_BATCH);
 		block_group->pinned = 0;
 
 		spin_unlock(&block_group->lock);
@@ -10782,8 +10667,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * Btrfs_remove_chunk will abort the transaction if things go
 		 * horribly wrong.
 		 */
-		ret = btrfs_remove_chunk(trans, fs_info,
-					 block_group->key.objectid);
+		ret = btrfs_remove_chunk(trans, block_group->key.objectid);
 
 		if (ret) {
 			if (trimming)
@@ -11066,3 +10950,16 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root)
 			       !atomic_read(&root->will_be_snapshotted));
 	}
 }
+
+void btrfs_mark_bg_unused(struct btrfs_block_group_cache *bg)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+
+	spin_lock(&fs_info->unused_bgs_lock);
+	if (list_empty(&bg->bg_list)) {
+		btrfs_get_block_group(bg);
+		trace_btrfs_add_unused_block_group(bg);
+		list_add_tail(&bg->bg_list, &fs_info->unused_bgs);
+	}
+	spin_unlock(&fs_info->unused_bgs_lock);
+}

fs/btrfs/extent_io.c

@@ -140,14 +140,6 @@ static int add_extent_changeset(struct extent_state *state, unsigned bits,
 
 static void flush_write_bio(struct extent_page_data *epd);
 
-static inline struct btrfs_fs_info *
-tree_fs_info(struct extent_io_tree *tree)
-{
-	if (tree->ops)
-		return tree->ops->tree_fs_info(tree->private_data);
-	return NULL;
-}
-
 int __init extent_io_init(void)
 {
 	extent_state_cache = kmem_cache_create("btrfs_extent_state",
@@ -564,8 +556,10 @@ alloc_extent_state_atomic(struct extent_state *prealloc)
 
 static void extent_io_tree_panic(struct extent_io_tree *tree, int err)
 {
-	btrfs_panic(tree_fs_info(tree), err,
-		    "Locking error: Extent tree was modified by another thread while locked.");
+	struct inode *inode = tree->private_data;
+
+	btrfs_panic(btrfs_sb(inode->i_sb), err,
+	"locking error: extent tree was modified by another thread while locked");
 }
 
 /*
@@ -1386,14 +1380,6 @@ void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end)
 	}
 }
 
-/*
- * helper function to set both pages and extents in the tree writeback
- */
-static void set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end)
-{
-	tree->ops->set_range_writeback(tree->private_data, start, end);
-}
-
 /* find the first state struct with 'bits' set after 'start', and
  * return it.  tree->lock must be held.  NULL will returned if
  * nothing was found after 'start'
@@ -2059,7 +2045,7 @@ int repair_eb_io_failure(struct btrfs_fs_info *fs_info,
 			 struct extent_buffer *eb, int mirror_num)
 {
 	u64 start = eb->start;
-	unsigned long i, num_pages = num_extent_pages(eb->start, eb->len);
+	int i, num_pages = num_extent_pages(eb);
 	int ret = 0;
 
 	if (sb_rdonly(fs_info->sb))
@@ -2398,7 +2384,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
 				      start - page_offset(page),
 				      (int)phy_offset, failed_bio->bi_end_io,
 				      NULL);
-	bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
+	bio->bi_opf = REQ_OP_READ | read_mode;
 
 	btrfs_debug(btrfs_sb(inode->i_sb),
 		"Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d",
@@ -2790,8 +2776,8 @@ static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree,
 		else
 			contig = bio_end_sector(bio) == sector;
 
-		if (tree->ops && tree->ops->merge_bio_hook(page, offset,
-					page_size, bio, bio_flags))
+		if (tree->ops && btrfs_merge_bio_hook(page, offset, page_size,
+						      bio, bio_flags))
 			can_merge = false;
 
 		if (prev_bio_flags != bio_flags || !contig || !can_merge ||
@@ -3422,7 +3408,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 			continue;
 		}
 
-		set_range_writeback(tree, cur, cur + iosize - 1);
+		btrfs_set_range_writeback(tree, cur, cur + iosize - 1);
 		if (!PageWriteback(page)) {
 			btrfs_err(BTRFS_I(inode)->root->fs_info,
 				   "page %lu not writeback, cur %llu end %llu",
@@ -3538,7 +3524,7 @@ lock_extent_buffer_for_io(struct extent_buffer *eb,
 			  struct btrfs_fs_info *fs_info,
 			  struct extent_page_data *epd)
 {
-	unsigned long i, num_pages;
+	int i, num_pages;
 	int flush = 0;
 	int ret = 0;
 
@@ -3588,7 +3574,7 @@ lock_extent_buffer_for_io(struct extent_buffer *eb,
 	if (!ret)
 		return ret;
 
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		struct page *p = eb->pages[i];
 
@@ -3712,13 +3698,13 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 	struct extent_io_tree *tree = &BTRFS_I(fs_info->btree_inode)->io_tree;
 	u64 offset = eb->start;
 	u32 nritems;
-	unsigned long i, num_pages;
+	int i, num_pages;
 	unsigned long start, end;
 	unsigned int write_flags = wbc_to_write_flags(wbc) | REQ_META;
 	int ret = 0;
 
 	clear_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags);
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	atomic_set(&eb->io_pages, num_pages);
 
 	/* set btree blocks beyond nritems with 0 to avoid stale content. */
@@ -4643,23 +4629,20 @@ int extent_buffer_under_io(struct extent_buffer *eb)
 }
 
 /*
- * Helper for releasing extent buffer page.
+ * Release all pages attached to the extent buffer.
  */
-static void btrfs_release_extent_buffer_page(struct extent_buffer *eb)
+static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb)
 {
-	unsigned long index;
-	struct page *page;
-	int mapped = !test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags);
+	int i;
+	int num_pages;
+	int mapped = !test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags);
 
 	BUG_ON(extent_buffer_under_io(eb));
 
-	index = num_extent_pages(eb->start, eb->len);
-	if (index == 0)
-		return;
+	num_pages = num_extent_pages(eb);
+	for (i = 0; i < num_pages; i++) {
+		struct page *page = eb->pages[i];
 
-	do {
-		index--;
-		page = eb->pages[index];
 		if (!page)
 			continue;
 		if (mapped)
@@ -4691,7 +4674,7 @@ static void btrfs_release_extent_buffer_page(struct extent_buffer *eb)
 
 		/* One for when we allocated the page */
 		put_page(page);
-	} while (index != 0);
+	}
 }
 
 /*
@@ -4699,7 +4682,7 @@ static void btrfs_release_extent_buffer_page(struct extent_buffer *eb)
  */
 static inline void btrfs_release_extent_buffer(struct extent_buffer *eb)
 {
-	btrfs_release_extent_buffer_page(eb);
+	btrfs_release_extent_buffer_pages(eb);
 	__free_extent_buffer(eb);
 }
 
@@ -4743,10 +4726,10 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start,
 
 struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src)
 {
-	unsigned long i;
+	int i;
 	struct page *p;
 	struct extent_buffer *new;
-	unsigned long num_pages = num_extent_pages(src->start, src->len);
+	int num_pages = num_extent_pages(src);
 
 	new = __alloc_extent_buffer(src->fs_info, src->start, src->len);
 	if (new == NULL)
@@ -4766,7 +4749,7 @@ struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src)
 	}
 
 	set_bit(EXTENT_BUFFER_UPTODATE, &new->bflags);
-	set_bit(EXTENT_BUFFER_DUMMY, &new->bflags);
+	set_bit(EXTENT_BUFFER_UNMAPPED, &new->bflags);
 
 	return new;
 }
@@ -4775,15 +4758,14 @@ struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 						  u64 start, unsigned long len)
 {
 	struct extent_buffer *eb;
-	unsigned long num_pages;
-	unsigned long i;
-
-	num_pages = num_extent_pages(start, len);
+	int num_pages;
+	int i;
 
 	eb = __alloc_extent_buffer(fs_info, start, len);
 	if (!eb)
 		return NULL;
 
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		eb->pages[i] = alloc_page(GFP_NOFS);
 		if (!eb->pages[i])
@@ -4791,7 +4773,7 @@ struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 	}
 	set_extent_buffer_uptodate(eb);
 	btrfs_set_header_nritems(eb, 0);
-	set_bit(EXTENT_BUFFER_DUMMY, &eb->bflags);
+	set_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags);
 
 	return eb;
 err:
@@ -4843,11 +4825,11 @@ static void check_buffer_tree_ref(struct extent_buffer *eb)
 static void mark_extent_buffer_accessed(struct extent_buffer *eb,
 		struct page *accessed)
 {
-	unsigned long num_pages, i;
+	int num_pages, i;
 
 	check_buffer_tree_ref(eb);
 
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		struct page *p = eb->pages[i];
 
@@ -4944,8 +4926,8 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 					  u64 start)
 {
 	unsigned long len = fs_info->nodesize;
-	unsigned long num_pages = num_extent_pages(start, len);
-	unsigned long i;
+	int num_pages;
+	int i;
 	unsigned long index = start >> PAGE_SHIFT;
 	struct extent_buffer *eb;
 	struct extent_buffer *exists = NULL;
@@ -4967,6 +4949,7 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 	if (!eb)
 		return ERR_PTR(-ENOMEM);
 
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++, index++) {
 		p = find_or_create_page(mapping, index, GFP_NOFS|__GFP_NOFAIL);
 		if (!p) {
@@ -5009,8 +4992,11 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 			uptodate = 0;
 
 		/*
-		 * see below about how we avoid a nasty race with release page
-		 * and why we unlock later
+		 * We can't unlock the pages just yet since the extent buffer
+		 * hasn't been properly inserted in the radix tree, this
+		 * opens a race with btree_releasepage which can free a page
+		 * while we are still filling in all pages for the buffer and
+		 * we could crash.
 		 */
 	}
 	if (uptodate)
@@ -5039,21 +5025,12 @@ again:
 	set_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags);
 
 	/*
-	 * there is a race where release page may have
-	 * tried to find this extent buffer in the radix
-	 * but failed.  It will tell the VM it is safe to
-	 * reclaim the, and it will clear the page private bit.
-	 * We must make sure to set the page private bit properly
-	 * after the extent buffer is in the radix tree so
-	 * it doesn't get lost
+	 * Now it's safe to unlock the pages because any calls to
+	 * btree_releasepage will correctly detect that a page belongs to a
+	 * live buffer and won't free them prematurely.
 	 */
-	SetPageChecked(eb->pages[0]);
-	for (i = 1; i < num_pages; i++) {
-		p = eb->pages[i];
-		ClearPageChecked(p);
-		unlock_page(p);
-	}
-	unlock_page(eb->pages[0]);
+	for (i = 0; i < num_pages; i++)
+		unlock_page(eb->pages[i]);
 	return eb;
 
 free_eb:
@@ -5075,9 +5052,10 @@ static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head)
 	__free_extent_buffer(eb);
 }
 
-/* Expects to have eb->eb_lock already held */
 static int release_extent_buffer(struct extent_buffer *eb)
 {
+	lockdep_assert_held(&eb->refs_lock);
+
 	WARN_ON(atomic_read(&eb->refs) == 0);
 	if (atomic_dec_and_test(&eb->refs)) {
 		if (test_and_clear_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags)) {
@@ -5094,9 +5072,9 @@ static int release_extent_buffer(struct extent_buffer *eb)
 		}
 
 		/* Should be safe to release our pages at this point */
-		btrfs_release_extent_buffer_page(eb);
+		btrfs_release_extent_buffer_pages(eb);
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-		if (unlikely(test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags))) {
+		if (unlikely(test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags))) {
 			__free_extent_buffer(eb);
 			return 1;
 		}
@@ -5127,7 +5105,7 @@ void free_extent_buffer(struct extent_buffer *eb)
 
 	spin_lock(&eb->refs_lock);
 	if (atomic_read(&eb->refs) == 2 &&
-	    test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags))
+	    test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags))
 		atomic_dec(&eb->refs);
 
 	if (atomic_read(&eb->refs) == 2 &&
@@ -5159,11 +5137,11 @@ void free_extent_buffer_stale(struct extent_buffer *eb)
 
 void clear_extent_buffer_dirty(struct extent_buffer *eb)
 {
-	unsigned long i;
-	unsigned long num_pages;
+	int i;
+	int num_pages;
 	struct page *page;
 
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 
 	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
@@ -5189,15 +5167,15 @@ void clear_extent_buffer_dirty(struct extent_buffer *eb)
 
 int set_extent_buffer_dirty(struct extent_buffer *eb)
 {
-	unsigned long i;
-	unsigned long num_pages;
+	int i;
+	int num_pages;
 	int was_dirty = 0;
 
 	check_buffer_tree_ref(eb);
 
 	was_dirty = test_and_set_bit(EXTENT_BUFFER_DIRTY, &eb->bflags);
 
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	WARN_ON(atomic_read(&eb->refs) == 0);
 	WARN_ON(!test_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags));
 
@@ -5208,12 +5186,12 @@ int set_extent_buffer_dirty(struct extent_buffer *eb)
 
 void clear_extent_buffer_uptodate(struct extent_buffer *eb)
 {
-	unsigned long i;
+	int i;
 	struct page *page;
-	unsigned long num_pages;
+	int num_pages;
 
 	clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
 		if (page)
@@ -5223,12 +5201,12 @@ void clear_extent_buffer_uptodate(struct extent_buffer *eb)
 
 void set_extent_buffer_uptodate(struct extent_buffer *eb)
 {
-	unsigned long i;
+	int i;
 	struct page *page;
-	unsigned long num_pages;
+	int num_pages;
 
 	set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
 		SetPageUptodate(page);
@@ -5238,13 +5216,13 @@ void set_extent_buffer_uptodate(struct extent_buffer *eb)
 int read_extent_buffer_pages(struct extent_io_tree *tree,
 			     struct extent_buffer *eb, int wait, int mirror_num)
 {
-	unsigned long i;
+	int i;
 	struct page *page;
 	int err;
 	int ret = 0;
 	int locked_pages = 0;
 	int all_uptodate = 1;
-	unsigned long num_pages;
+	int num_pages;
 	unsigned long num_reads = 0;
 	struct bio *bio = NULL;
 	unsigned long bio_flags = 0;
@@ -5252,7 +5230,7 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))
 		return 0;
 
-	num_pages = num_extent_pages(eb->start, eb->len);
+	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
 		if (wait == WAIT_NONE) {
@@ -5576,11 +5554,11 @@ void copy_extent_buffer_full(struct extent_buffer *dst,
 			     struct extent_buffer *src)
 {
 	int i;
-	unsigned num_pages;
+	int num_pages;
 
 	ASSERT(dst->len == src->len);
 
-	num_pages = num_extent_pages(dst->start, dst->len);
+	num_pages = num_extent_pages(dst);
 	for (i = 0; i < num_pages; i++)
 		copy_page(page_address(dst->pages[i]),
 				page_address(src->pages[i]));

fs/btrfs/extent_io.h

@@ -46,7 +46,7 @@
 #define EXTENT_BUFFER_STALE 6
 #define EXTENT_BUFFER_WRITEBACK 7
 #define EXTENT_BUFFER_READ_ERR 8        /* read IO error */
-#define EXTENT_BUFFER_DUMMY 9
+#define EXTENT_BUFFER_UNMAPPED 9
 #define EXTENT_BUFFER_IN_TREE 10
 #define EXTENT_BUFFER_WRITE_ERR 11    /* write IO error */
 
@@ -92,9 +92,6 @@ typedef	blk_status_t (extent_submit_bio_hook_t)(void *private_data, struct bio *
 typedef blk_status_t (extent_submit_bio_start_t)(void *private_data,
 		struct bio *bio, u64 bio_offset);
 
-typedef blk_status_t (extent_submit_bio_done_t)(void *private_data,
-		struct bio *bio, int mirror_num);
-
 struct extent_io_ops {
 	/*
 	 * The following callbacks must be allways defined, the function
@@ -104,12 +101,7 @@ struct extent_io_ops {
 	int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
 				    struct page *page, u64 start, u64 end,
 				    int mirror);
-	int (*merge_bio_hook)(struct page *page, unsigned long offset,
-			      size_t size, struct bio *bio,
-			      unsigned long bio_flags);
 	int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
-	struct btrfs_fs_info *(*tree_fs_info)(void *private_data);
-	void (*set_range_writeback)(void *private_data, u64 start, u64 end);
 
 	/*
 	 * Optional hooks, called if the pointer is not NULL
@@ -440,10 +432,10 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 			     int mirror_num);
 void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
 
-static inline unsigned long num_extent_pages(u64 start, u64 len)
+static inline int num_extent_pages(const struct extent_buffer *eb)
 {
-	return ((start + len + PAGE_SIZE - 1) >> PAGE_SHIFT) -
-		(start >> PAGE_SHIFT);
+	return (round_up(eb->start + eb->len, PAGE_SIZE) >> PAGE_SHIFT) -
+	       (eb->start >> PAGE_SHIFT);
 }
 
 static inline void extent_buffer_get(struct extent_buffer *eb)

fs/btrfs/file-item.c

@@ -922,7 +922,7 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 				     const bool new_inline,
 				     struct extent_map *em)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_root *root = inode->root;
 	struct extent_buffer *leaf = path->nodes[0];
 	const int slot = path->slots[0];
@@ -942,7 +942,7 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 			btrfs_file_extent_num_bytes(leaf, fi);
 	} else if (type == BTRFS_FILE_EXTENT_INLINE) {
 		size_t size;
-		size = btrfs_file_extent_inline_len(leaf, slot, fi);
+		size = btrfs_file_extent_ram_bytes(leaf, fi);
 		extent_end = ALIGN(extent_start + size,
 				   fs_info->sectorsize);
 	}

fs/btrfs/file.c

@@ -5,14 +5,11 @@
 
 #include <linux/fs.h>
 #include <linux/pagemap.h>
-#include <linux/highmem.h>
 #include <linux/time.h>
 #include <linux/init.h>
 #include <linux/string.h>
 #include <linux/backing-dev.h>
-#include <linux/mpage.h>
 #include <linux/falloc.h>
-#include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/compat.h>
 #include <linux/slab.h>
@@ -83,7 +80,7 @@ static int __compare_inode_defrag(struct inode_defrag *defrag1,
 static int __btrfs_add_inode_defrag(struct btrfs_inode *inode,
 				    struct inode_defrag *defrag)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct inode_defrag *entry;
 	struct rb_node **p;
 	struct rb_node *parent = NULL;
@@ -135,8 +132,8 @@ static inline int __need_auto_defrag(struct btrfs_fs_info *fs_info)
 int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
 			   struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
 	struct btrfs_root *root = inode->root;
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct inode_defrag *defrag;
 	u64 transid;
 	int ret;
@@ -185,7 +182,7 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
 static void btrfs_requeue_inode_defrag(struct btrfs_inode *inode,
 				       struct inode_defrag *defrag)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	int ret;
 
 	if (!__need_auto_defrag(fs_info))
@@ -833,8 +830,7 @@ next_slot:
 				btrfs_file_extent_num_bytes(leaf, fi);
 		} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 			extent_end = key.offset +
-				btrfs_file_extent_inline_len(leaf,
-						     path->slots[0], fi);
+				btrfs_file_extent_ram_bytes(leaf, fi);
 		} else {
 			/* can't happen */
 			BUG();
@@ -1133,7 +1129,7 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot,
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 			      struct btrfs_inode *inode, u64 start, u64 end)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = inode->root;
 	struct extent_buffer *leaf;
 	struct btrfs_path *path;
@@ -1470,7 +1466,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 				u64 *lockstart, u64 *lockend,
 				struct extent_state **cached_state)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	u64 start_pos;
 	u64 last_pos;
 	int i;
@@ -1526,7 +1522,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
 				    size_t *write_bytes)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_root *root = inode->root;
 	struct btrfs_ordered_extent *ordered;
 	u64 lockstart, lockend;
@@ -1569,10 +1565,11 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
 	return ret;
 }
 
-static noinline ssize_t __btrfs_buffered_write(struct file *file,
-					       struct iov_iter *i,
-					       loff_t pos)
+static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
+					       struct iov_iter *i)
 {
+	struct file *file = iocb->ki_filp;
+	loff_t pos = iocb->ki_pos;
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -1804,7 +1801,7 @@ static ssize_t __btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
-	loff_t pos = iocb->ki_pos;
+	loff_t pos;
 	ssize_t written;
 	ssize_t written_buffered;
 	loff_t endbyte;
@@ -1815,8 +1812,8 @@ static ssize_t __btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	if (written < 0 || !iov_iter_count(from))
 		return written;
 
-	pos += written;
-	written_buffered = __btrfs_buffered_write(file, from, pos);
+	pos = iocb->ki_pos;
+	written_buffered = btrfs_buffered_write(iocb, from);
 	if (written_buffered < 0) {
 		err = written_buffered;
 		goto out;
@@ -1953,7 +1950,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	if (iocb->ki_flags & IOCB_DIRECT) {
 		num_written = __btrfs_direct_write(iocb, from);
 	} else {
-		num_written = __btrfs_buffered_write(file, from, pos);
+		num_written = btrfs_buffered_write(iocb, from);
 		if (num_written > 0)
 			iocb->ki_pos = pos + num_written;
 		if (clean_page)
@@ -2042,7 +2039,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_log_ctx ctx;
 	int ret = 0, err;
-	bool full_sync = false;
 	u64 len;
 
 	/*
@@ -2066,96 +2062,21 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 
 	inode_lock(inode);
 	atomic_inc(&root->log_batch);
-	full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
-			     &BTRFS_I(inode)->runtime_flags);
+
 	/*
-	 * We might have have had more pages made dirty after calling
-	 * start_ordered_ops and before acquiring the inode's i_mutex.
+	 * We have to do this here to avoid the priority inversion of waiting on
+	 * IO of a lower priority task while holding a transaciton open.
 	 */
-	if (full_sync) {
-		/*
-		 * For a full sync, we need to make sure any ordered operations
-		 * start and finish before we start logging the inode, so that
-		 * all extents are persisted and the respective file extent
-		 * items are in the fs/subvol btree.
-		 */
-		ret = btrfs_wait_ordered_range(inode, start, len);
-	} else {
-		/*
-		 * Start any new ordered operations before starting to log the
-		 * inode. We will wait for them to finish in btrfs_sync_log().
-		 *
-		 * Right before acquiring the inode's mutex, we might have new
-		 * writes dirtying pages, which won't immediately start the
-		 * respective ordered operations - that is done through the
-		 * fill_delalloc callbacks invoked from the writepage and
-		 * writepages address space operations. So make sure we start
-		 * all ordered operations before starting to log our inode. Not
-		 * doing this means that while logging the inode, writeback
-		 * could start and invoke writepage/writepages, which would call
-		 * the fill_delalloc callbacks (cow_file_range,
-		 * submit_compressed_extents). These callbacks add first an
-		 * extent map to the modified list of extents and then create
-		 * the respective ordered operation, which means in
-		 * tree-log.c:btrfs_log_inode() we might capture all existing
-		 * ordered operations (with btrfs_get_logged_extents()) before
-		 * the fill_delalloc callback adds its ordered operation, and by
-		 * the time we visit the modified list of extent maps (with
-		 * btrfs_log_changed_extents()), we see and process the extent
-		 * map they created. We then use the extent map to construct a
-		 * file extent item for logging without waiting for the
-		 * respective ordered operation to finish - this file extent
-		 * item points to a disk location that might not have yet been
-		 * written to, containing random data - so after a crash a log
-		 * replay will make our inode have file extent items that point
-		 * to disk locations containing invalid data, as we returned
-		 * success to userspace without waiting for the respective
-		 * ordered operation to finish, because it wasn't captured by
-		 * btrfs_get_logged_extents().
-		 */
-		ret = start_ordered_ops(inode, start, end);
-	}
+	ret = btrfs_wait_ordered_range(inode, start, len);
 	if (ret) {
 		inode_unlock(inode);
 		goto out;
 	}
 	atomic_inc(&root->log_batch);
 
-	/*
-	 * If the last transaction that changed this file was before the current
-	 * transaction and we have the full sync flag set in our inode, we can
-	 * bail out now without any syncing.
-	 *
-	 * Note that we can't bail out if the full sync flag isn't set. This is
-	 * because when the full sync flag is set we start all ordered extents
-	 * and wait for them to fully complete - when they complete they update
-	 * the inode's last_trans field through:
-	 *
-	 *     btrfs_finish_ordered_io() ->
-	 *         btrfs_update_inode_fallback() ->
-	 *             btrfs_update_inode() ->
-	 *                 btrfs_set_inode_last_trans()
-	 *
-	 * So we are sure that last_trans is up to date and can do this check to
-	 * bail out safely. For the fast path, when the full sync flag is not
-	 * set in our inode, we can not do it because we start only our ordered
-	 * extents and don't wait for them to complete (that is when
-	 * btrfs_finish_ordered_io runs), so here at this point their last_trans
-	 * value might be less than or equals to fs_info->last_trans_committed,
-	 * and setting a speculative last_trans for an inode when a buffered
-	 * write is made (such as fs_info->generation + 1 for example) would not
-	 * be reliable since after setting the value and before fsync is called
-	 * any number of transactions can start and commit (transaction kthread
-	 * commits the current transaction periodically), and a transaction
-	 * commit does not start nor waits for ordered extents to complete.
-	 */
 	smp_mb();
 	if (btrfs_inode_in_log(BTRFS_I(inode), fs_info->generation) ||
-	    (full_sync && BTRFS_I(inode)->last_trans <=
-	     fs_info->last_trans_committed) ||
-	    (!btrfs_have_ordered_extents_in_range(inode, start, len) &&
-	     BTRFS_I(inode)->last_trans
-	     <= fs_info->last_trans_committed)) {
+	    BTRFS_I(inode)->last_trans <= fs_info->last_trans_committed) {
 		/*
 		 * We've had everything committed since the last time we were
 		 * modified so clear this flag in case it was set for whatever
@@ -2239,13 +2160,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 				goto out;
 			}
 		}
-		if (!full_sync) {
-			ret = btrfs_wait_ordered_range(inode, start, len);
-			if (ret) {
-				btrfs_end_transaction(trans);
-				goto out;
-			}
-		}
 		ret = btrfs_commit_transaction(trans);
 	} else {
 		ret = btrfs_end_transaction(trans);
@@ -2310,7 +2224,7 @@ static int fill_holes(struct btrfs_trans_handle *trans,
 		struct btrfs_inode *inode,
 		struct btrfs_path *path, u64 offset, u64 end)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = inode->root;
 	struct extent_buffer *leaf;
 	struct btrfs_file_extent_item *fi;

fs/btrfs/free-space-cache.c

@@ -71,10 +71,6 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	inode = btrfs_iget(fs_info->sb, &location, root, NULL);
 	if (IS_ERR(inode))
 		return inode;
-	if (is_bad_inode(inode)) {
-		iput(inode);
-		return ERR_PTR(-ENOENT);
-	}
 
 	mapping_set_gfp_mask(inode->i_mapping,
 			mapping_gfp_constraint(inode->i_mapping,
@@ -300,9 +296,9 @@ static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode,
 	if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FREE_INO_OBJECTID)
 		check_crcs = 1;
 
-	/* Make sure we can fit our crcs into the first page */
+	/* Make sure we can fit our crcs and generation into the first page */
 	if (write && check_crcs &&
-	    (num_pages * sizeof(u32)) >= PAGE_SIZE)
+	    (num_pages * sizeof(u32) + sizeof(u64)) > PAGE_SIZE)
 		return -ENOSPC;
 
 	memset(io_ctl, 0, sizeof(struct btrfs_io_ctl));
@@ -547,7 +543,7 @@ static int io_ctl_add_bitmap(struct btrfs_io_ctl *io_ctl, void *bitmap)
 		io_ctl_map_page(io_ctl, 0);
 	}
 
-	memcpy(io_ctl->cur, bitmap, PAGE_SIZE);
+	copy_page(io_ctl->cur, bitmap);
 	io_ctl_set_crc(io_ctl, io_ctl->index - 1);
 	if (io_ctl->index < io_ctl->num_pages)
 		io_ctl_map_page(io_ctl, 0);
@@ -607,7 +603,7 @@ static int io_ctl_read_bitmap(struct btrfs_io_ctl *io_ctl,
 	if (ret)
 		return ret;
 
-	memcpy(entry->bitmap, io_ctl->cur, PAGE_SIZE);
+	copy_page(entry->bitmap, io_ctl->cur);
 	io_ctl_unmap_page(io_ctl);
 
 	return 0;
@@ -655,7 +651,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 				   struct btrfs_free_space_ctl *ctl,
 				   struct btrfs_path *path, u64 offset)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_free_space_header *header;
 	struct extent_buffer *leaf;
 	struct btrfs_io_ctl io_ctl;
@@ -1123,13 +1119,10 @@ static int __btrfs_wait_cache_io(struct btrfs_root *root,
 {
 	int ret;
 	struct inode *inode = io_ctl->inode;
-	struct btrfs_fs_info *fs_info;
 
 	if (!inode)
 		return 0;
 
-	fs_info = btrfs_sb(inode->i_sb);
-
 	/* Flush the dirty pages in the cache file. */
 	ret = flush_dirty_cache(inode);
 	if (ret)
@@ -1145,7 +1138,7 @@ out:
 		BTRFS_I(inode)->generation = 0;
 		if (block_group) {
 #ifdef DEBUG
-			btrfs_err(fs_info,
+			btrfs_err(root->fs_info,
 				  "failed to write free space cache for block group %llu",
 				  block_group->key.objectid);
 #endif

fs/btrfs/free-space-tree.c

@@ -1236,7 +1236,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
 	if (ret)
 		goto abort;
 
-	ret = btrfs_del_root(trans, fs_info, &free_space_root->root_key);
+	ret = btrfs_del_root(trans, &free_space_root->root_key);
 	if (ret)
 		goto abort;
 

fs/btrfs/inode-map.c

@@ -3,7 +3,6 @@
  * Copyright (C) 2007 Oracle.  All rights reserved.
  */
 
-#include <linux/delay.h>
 #include <linux/kthread.h>
 #include <linux/pagemap.h>
 
@@ -244,8 +243,6 @@ void btrfs_unpin_free_ino(struct btrfs_root *root)
 		return;
 
 	while (1) {
-		bool add_to_ctl = true;
-
 		spin_lock(rbroot_lock);
 		n = rb_first(rbroot);
 		if (!n) {
@@ -257,15 +254,14 @@ void btrfs_unpin_free_ino(struct btrfs_root *root)
 		BUG_ON(info->bitmap); /* Logic error */
 
 		if (info->offset > root->ino_cache_progress)
-			add_to_ctl = false;
-		else if (info->offset + info->bytes > root->ino_cache_progress)
-			count = root->ino_cache_progress - info->offset + 1;
+			count = 0;
 		else
-			count = info->bytes;
+			count = min(root->ino_cache_progress - info->offset + 1,
+				    info->bytes);
 
 		rb_erase(&info->offset_index, rbroot);
 		spin_unlock(rbroot_lock);
-		if (add_to_ctl)
+		if (count)
 			__btrfs_add_free_space(root->fs_info, ctl,
 					       info->offset, count);
 		kmem_cache_free(btrfs_free_space_cachep, info);

fs/btrfs/inode.c

@@ -14,17 +14,13 @@
 #include <linux/init.h>
 #include <linux/string.h>
 #include <linux/backing-dev.h>
-#include <linux/mpage.h>
-#include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/compat.h>
-#include <linux/bit_spinlock.h>
 #include <linux/xattr.h>
 #include <linux/posix_acl.h>
 #include <linux/falloc.h>
 #include <linux/slab.h>
 #include <linux/ratelimit.h>
-#include <linux/mount.h>
 #include <linux/btrfs.h>
 #include <linux/blkdev.h>
 #include <linux/posix_acl_xattr.h>
@@ -1443,8 +1439,7 @@ next_slot:
 			nocow = 1;
 		} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 			extent_end = found_key.offset +
-				btrfs_file_extent_inline_len(leaf,
-						     path->slots[0], fi);
+				btrfs_file_extent_ram_bytes(leaf, fi);
 			extent_end = ALIGN(extent_end,
 					   fs_info->sectorsize);
 		} else {
@@ -1752,7 +1747,7 @@ static void btrfs_add_delalloc_inodes(struct btrfs_root *root,
 void __btrfs_del_delalloc_inode(struct btrfs_root *root,
 				struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = root->fs_info;
 
 	if (!list_empty(&inode->delalloc_inodes)) {
 		list_del_init(&inode->delalloc_inodes);
@@ -1903,8 +1898,8 @@ static void btrfs_clear_bit_hook(void *private_data,
 }
 
 /*
- * extent_io.c merge_bio_hook, this must check the chunk tree to make sure
- * we don't create bios that span stripes or chunks
+ * Merge bio hook, this must check the chunk tree to make sure we don't create
+ * bios that span stripes or chunks
  *
  * return 1 if page cannot be merged to bio
  * return 0 if page can be merged to bio
@@ -1962,7 +1957,7 @@ static blk_status_t btrfs_submit_bio_start(void *private_data, struct bio *bio,
  * At IO completion time the cums attached on the ordered extent record
  * are inserted into the btree
  */
-static blk_status_t btrfs_submit_bio_done(void *private_data, struct bio *bio,
+blk_status_t btrfs_submit_bio_done(void *private_data, struct bio *bio,
 			  int mirror_num)
 {
 	struct inode *inode = private_data;
@@ -2035,8 +2030,7 @@ static blk_status_t btrfs_submit_bio_hook(void *private_data, struct bio *bio,
 		/* we're doing a write, do the async checksumming */
 		ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, bio_flags,
 					  bio_offset, inode,
-					  btrfs_submit_bio_start,
-					  btrfs_submit_bio_done);
+					  btrfs_submit_bio_start);
 		goto out;
 	} else if (!skip_sum) {
 		ret = btrfs_csum_one_bio(inode, bio, 0, 0);
@@ -3610,18 +3604,15 @@ static int btrfs_read_locked_inode(struct inode *inode)
 		filled = true;
 
 	path = btrfs_alloc_path();
-	if (!path) {
-		ret = -ENOMEM;
-		goto make_bad;
-	}
+	if (!path)
+		return -ENOMEM;
 
 	memcpy(&location, &BTRFS_I(inode)->location, sizeof(location));
 
 	ret = btrfs_lookup_inode(NULL, root, path, &location, 0);
 	if (ret) {
-		if (ret > 0)
-			ret = -ENOENT;
-		goto make_bad;
+		btrfs_free_path(path);
+		return ret;
 	}
 
 	leaf = path->nodes[0];
@@ -3774,11 +3765,6 @@ cache_acl:
 
 	btrfs_sync_inode_flags_to_i_flags(inode);
 	return 0;
-
-make_bad:
-	btrfs_free_path(path);
-	make_bad_inode(inode);
-	return ret;
 }
 
 /*
@@ -3984,7 +3970,7 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 		goto err;
 	}
 skip_backref:
-	ret = btrfs_delete_delayed_dir_index(trans, fs_info, dir, index);
+	ret = btrfs_delete_delayed_dir_index(trans, dir, index);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto err;
@@ -4087,11 +4073,10 @@ out:
 }
 
 static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
-			struct btrfs_root *root,
-			struct inode *dir, u64 objectid,
-			const char *name, int name_len)
+			       struct inode *dir, u64 objectid,
+			       const char *name, int name_len)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_path *path;
 	struct extent_buffer *leaf;
 	struct btrfs_dir_item *di;
@@ -4124,9 +4109,8 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	}
 	btrfs_release_path(path);
 
-	ret = btrfs_del_root_ref(trans, fs_info, objectid,
-				 root->root_key.objectid, dir_ino,
-				 &index, name, name_len);
+	ret = btrfs_del_root_ref(trans, objectid, root->root_key.objectid,
+				 dir_ino, &index, name, name_len);
 	if (ret < 0) {
 		if (ret != -ENOENT) {
 			btrfs_abort_transaction(trans, ret);
@@ -4145,12 +4129,11 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 
 		leaf = path->nodes[0];
 		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-		btrfs_release_path(path);
 		index = key.offset;
 	}
 	btrfs_release_path(path);
 
-	ret = btrfs_delete_delayed_dir_index(trans, fs_info, BTRFS_I(dir), index);
+	ret = btrfs_delete_delayed_dir_index(trans, BTRFS_I(dir), index);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
@@ -4243,9 +4226,9 @@ again:
 		prev = node;
 		entry = rb_entry(node, struct btrfs_inode, rb_node);
 
-		if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode)))
+		if (objectid < btrfs_ino(entry))
 			node = node->rb_left;
-		else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode)))
+		else if (objectid > btrfs_ino(entry))
 			node = node->rb_right;
 		else
 			break;
@@ -4253,7 +4236,7 @@ again:
 	if (!node) {
 		while (prev) {
 			entry = rb_entry(prev, struct btrfs_inode, rb_node);
-			if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) {
+			if (objectid <= btrfs_ino(entry)) {
 				node = prev;
 				break;
 			}
@@ -4262,7 +4245,7 @@ again:
 	}
 	while (node) {
 		entry = rb_entry(node, struct btrfs_inode, rb_node);
-		objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1;
+		objectid = btrfs_ino(entry) + 1;
 		inode = igrab(&entry->vfs_inode);
 		if (inode) {
 			spin_unlock(&root->inode_lock);
@@ -4343,10 +4326,8 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry)
 
 	btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
 
-	ret = btrfs_unlink_subvol(trans, root, dir,
-				dest->root_key.objectid,
-				dentry->d_name.name,
-				dentry->d_name.len);
+	ret = btrfs_unlink_subvol(trans, dir, dest->root_key.objectid,
+				  dentry->d_name.name, dentry->d_name.len);
 	if (ret) {
 		err = ret;
 		btrfs_abort_transaction(trans, ret);
@@ -4441,7 +4422,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 		return PTR_ERR(trans);
 
 	if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
-		err = btrfs_unlink_subvol(trans, root, dir,
+		err = btrfs_unlink_subvol(trans, dir,
 					  BTRFS_I(inode)->location.objectid,
 					  dentry->d_name.name,
 					  dentry->d_name.len);
@@ -4643,8 +4624,8 @@ search_again:
 					BTRFS_I(inode), leaf, fi,
 					found_key.offset);
 			} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
-				item_end += btrfs_file_extent_inline_len(leaf,
-							 path->slots[0], fi);
+				item_end += btrfs_file_extent_ram_bytes(leaf,
+									fi);
 
 				trace_btrfs_truncate_show_fi_inline(
 					BTRFS_I(inode), leaf, fi, path->slots[0],
@@ -5615,9 +5596,9 @@ static void inode_tree_add(struct inode *inode)
 		parent = *p;
 		entry = rb_entry(parent, struct btrfs_inode, rb_node);
 
-		if (ino < btrfs_ino(BTRFS_I(&entry->vfs_inode)))
+		if (ino < btrfs_ino(entry))
 			p = &parent->rb_left;
-		else if (ino > btrfs_ino(BTRFS_I(&entry->vfs_inode)))
+		else if (ino > btrfs_ino(entry))
 			p = &parent->rb_right;
 		else {
 			WARN_ON(!(entry->vfs_inode.i_state &
@@ -5708,16 +5689,21 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
 		int ret;
 
 		ret = btrfs_read_locked_inode(inode);
-		if (!is_bad_inode(inode)) {
+		if (!ret) {
 			inode_tree_add(inode);
 			unlock_new_inode(inode);
 			if (new)
 				*new = 1;
 		} else {
-			unlock_new_inode(inode);
-			iput(inode);
-			ASSERT(ret < 0);
-			inode = ERR_PTR(ret < 0 ? ret : -ESTALE);
+			iget_failed(inode);
+			/*
+			 * ret > 0 can come from btrfs_search_slot called by
+			 * btrfs_read_locked_inode, this means the inode item
+			 * was not found.
+			 */
+			if (ret > 0)
+				ret = -ENOENT;
+			inode = ERR_PTR(ret);
 		}
 	}
 
@@ -5745,7 +5731,7 @@ static struct inode *new_simple_dir(struct super_block *s,
 	inode->i_mtime = current_time(inode);
 	inode->i_atime = inode->i_mtime;
 	inode->i_ctime = inode->i_mtime;
-	BTRFS_I(inode)->i_otime = timespec64_to_timespec(inode->i_mtime);
+	BTRFS_I(inode)->i_otime = inode->i_mtime;
 
 	return inode;
 }
@@ -6027,32 +6013,6 @@ err:
 	return ret;
 }
 
-int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc)
-{
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_trans_handle *trans;
-	int ret = 0;
-	bool nolock = false;
-
-	if (test_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags))
-		return 0;
-
-	if (btrfs_fs_closing(root->fs_info) &&
-			btrfs_is_free_space_inode(BTRFS_I(inode)))
-		nolock = true;
-
-	if (wbc->sync_mode == WB_SYNC_ALL) {
-		if (nolock)
-			trans = btrfs_join_transaction_nolock(root);
-		else
-			trans = btrfs_join_transaction(root);
-		if (IS_ERR(trans))
-			return PTR_ERR(trans);
-		ret = btrfs_commit_transaction(trans);
-	}
-	return ret;
-}
-
 /*
  * This is somewhat expensive, updating the tree every time the
  * inode changes.  But, it is most likely to find the inode in cache.
@@ -6351,7 +6311,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	inode->i_mtime = current_time(inode);
 	inode->i_atime = inode->i_mtime;
 	inode->i_ctime = inode->i_mtime;
-	BTRFS_I(inode)->i_otime = timespec64_to_timespec(inode->i_mtime);
+	BTRFS_I(inode)->i_otime = inode->i_mtime;
 
 	inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
 				  struct btrfs_inode_item);
@@ -6420,7 +6380,6 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 		   struct btrfs_inode *parent_inode, struct btrfs_inode *inode,
 		   const char *name, int name_len, int add_backref, u64 index)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
 	int ret = 0;
 	struct btrfs_key key;
 	struct btrfs_root *root = parent_inode->root;
@@ -6436,7 +6395,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 	}
 
 	if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) {
-		ret = btrfs_add_root_ref(trans, fs_info, key.objectid,
+		ret = btrfs_add_root_ref(trans, key.objectid,
 					 root->root_key.objectid, parent_ino,
 					 index, name, name_len);
 	} else if (add_backref) {
@@ -6472,7 +6431,7 @@ fail_dir_item:
 	if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) {
 		u64 local_index;
 		int err;
-		err = btrfs_del_root_ref(trans, fs_info, key.objectid,
+		err = btrfs_del_root_ref(trans, key.objectid,
 					 root->root_key.objectid, parent_ino,
 					 &local_index, name, name_len);
 
@@ -6832,7 +6791,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	    size_t pg_offset, u64 start, u64 len,
 		int create)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	int ret;
 	int err = 0;
 	u64 extent_start = 0;
@@ -6928,7 +6887,8 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 						       extent_start);
 	} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
 		size_t size;
-		size = btrfs_file_extent_inline_len(leaf, path->slots[0], item);
+
+		size = btrfs_file_extent_ram_bytes(leaf, item);
 		extent_end = ALIGN(extent_start + size,
 				   fs_info->sectorsize);
 
@@ -6979,7 +6939,7 @@ next:
 		if (new_inline)
 			goto out;
 
-		size = btrfs_file_extent_inline_len(leaf, path->slots[0], item);
+		size = btrfs_file_extent_ram_bytes(leaf, item);
 		extent_offset = page_offset(page) + pg_offset - extent_start;
 		copy_size = min_t(u64, PAGE_SIZE - pg_offset,
 				  size - extent_offset);
@@ -7850,7 +7810,7 @@ static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio,
 	isector >>= inode->i_sb->s_blocksize_bits;
 	bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page,
 				pgoff, isector, repair_endio, repair_arg);
-	bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
+	bio->bi_opf = REQ_OP_READ | read_mode;
 
 	btrfs_debug(BTRFS_I(inode)->root->fs_info,
 		    "repair DIO read error: submitting new dio read[%#x] to this_mirror=%d, in_validation=%d",
@@ -8284,8 +8244,7 @@ static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio,
 	if (write && async_submit) {
 		ret = btrfs_wq_submit_bio(fs_info, bio, 0, 0,
 					  file_offset, inode,
-					  btrfs_submit_bio_start_direct_io,
-					  btrfs_submit_bio_done);
+					  btrfs_submit_bio_start_direct_io);
 		goto err;
 	} else if (write) {
 		/*
@@ -9525,8 +9484,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	/* src is a subvolume */
 	if (old_ino == BTRFS_FIRST_FREE_OBJECTID) {
 		root_objectid = BTRFS_I(old_inode)->root->root_key.objectid;
-		ret = btrfs_unlink_subvol(trans, root, old_dir,
-					  root_objectid,
+		ret = btrfs_unlink_subvol(trans, old_dir, root_objectid,
 					  old_dentry->d_name.name,
 					  old_dentry->d_name.len);
 	} else { /* src is an inode */
@@ -9545,8 +9503,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	/* dest is a subvolume */
 	if (new_ino == BTRFS_FIRST_FREE_OBJECTID) {
 		root_objectid = BTRFS_I(new_inode)->root->root_key.objectid;
-		ret = btrfs_unlink_subvol(trans, dest, new_dir,
-					  root_objectid,
+		ret = btrfs_unlink_subvol(trans, new_dir, root_objectid,
 					  new_dentry->d_name.name,
 					  new_dentry->d_name.len);
 	} else { /* dest is an inode */
@@ -9806,7 +9763,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) {
 		root_objectid = BTRFS_I(old_inode)->root->root_key.objectid;
-		ret = btrfs_unlink_subvol(trans, root, old_dir, root_objectid,
+		ret = btrfs_unlink_subvol(trans, old_dir, root_objectid,
 					old_dentry->d_name.name,
 					old_dentry->d_name.len);
 	} else {
@@ -9828,8 +9785,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (unlikely(btrfs_ino(BTRFS_I(new_inode)) ==
 			     BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
 			root_objectid = BTRFS_I(new_inode)->location.objectid;
-			ret = btrfs_unlink_subvol(trans, dest, new_dir,
-						root_objectid,
+			ret = btrfs_unlink_subvol(trans, new_dir, root_objectid,
 						new_dentry->d_name.name,
 						new_dentry->d_name.len);
 			BUG_ON(new_inode->i_nlink == 0);
@@ -10451,12 +10407,6 @@ static int btrfs_readpage_io_failed_hook(struct page *page, int failed_mirror)
 	return -EAGAIN;
 }
 
-static struct btrfs_fs_info *iotree_fs_info(void *private_data)
-{
-	struct inode *inode = private_data;
-	return btrfs_sb(inode->i_sb);
-}
-
 static void btrfs_check_extent_io_range(void *private_data, const char *caller,
 					u64 start, u64 end)
 {
@@ -10471,9 +10421,9 @@ static void btrfs_check_extent_io_range(void *private_data, const char *caller,
 	}
 }
 
-void btrfs_set_range_writeback(void *private_data, u64 start, u64 end)
+void btrfs_set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end)
 {
-	struct inode *inode = private_data;
+	struct inode *inode = tree->private_data;
 	unsigned long index = start >> PAGE_SHIFT;
 	unsigned long end_index = end >> PAGE_SHIFT;
 	struct page *page;
@@ -10529,10 +10479,7 @@ static const struct extent_io_ops btrfs_extent_io_ops = {
 	/* mandatory callbacks */
 	.submit_bio_hook = btrfs_submit_bio_hook,
 	.readpage_end_io_hook = btrfs_readpage_end_io_hook,
-	.merge_bio_hook = btrfs_merge_bio_hook,
 	.readpage_io_failed_hook = btrfs_readpage_io_failed_hook,
-	.tree_fs_info = iotree_fs_info,
-	.set_range_writeback = btrfs_set_range_writeback,
 
 	/* optional callbacks */
 	.fill_delalloc = run_delalloc_range,

fs/btrfs/ioctl.c

@@ -5,23 +5,18 @@
 
 #include <linux/kernel.h>
 #include <linux/bio.h>
-#include <linux/buffer_head.h>
 #include <linux/file.h>
 #include <linux/fs.h>
 #include <linux/fsnotify.h>
 #include <linux/pagemap.h>
 #include <linux/highmem.h>
 #include <linux/time.h>
-#include <linux/init.h>
 #include <linux/string.h>
 #include <linux/backing-dev.h>
 #include <linux/mount.h>
-#include <linux/mpage.h>
 #include <linux/namei.h>
-#include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/compat.h>
-#include <linux/bit_spinlock.h>
 #include <linux/security.h>
 #include <linux/xattr.h>
 #include <linux/mm.h>
@@ -606,7 +601,7 @@ static noinline int create_subvol(struct inode *dir,
 	trans->block_rsv = &block_rsv;
 	trans->bytes_reserved = block_rsv.size;
 
-	ret = btrfs_qgroup_inherit(trans, fs_info, 0, objectid, inherit);
+	ret = btrfs_qgroup_inherit(trans, 0, objectid, inherit);
 	if (ret)
 		goto fail;
 
@@ -616,14 +611,6 @@ static noinline int create_subvol(struct inode *dir,
 		goto fail;
 	}
 
-	memzero_extent_buffer(leaf, 0, sizeof(struct btrfs_header));
-	btrfs_set_header_bytenr(leaf, leaf->start);
-	btrfs_set_header_generation(leaf, trans->transid);
-	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
-	btrfs_set_header_owner(leaf, objectid);
-
-	write_extent_buffer_fsid(leaf, fs_info->fsid);
-	write_extent_buffer_chunk_tree_uuid(leaf, fs_info->chunk_tree_uuid);
 	btrfs_mark_buffer_dirty(leaf);
 
 	inode_item = &root_item->inode;
@@ -711,8 +698,7 @@ static noinline int create_subvol(struct inode *dir,
 	ret = btrfs_update_inode(trans, root, dir);
 	BUG_ON(ret);
 
-	ret = btrfs_add_root_ref(trans, fs_info,
-				 objectid, root->root_key.objectid,
+	ret = btrfs_add_root_ref(trans, objectid, root->root_key.objectid,
 				 btrfs_ino(BTRFS_I(dir)), index, name, namelen);
 	BUG_ON(ret);
 
@@ -2507,8 +2493,8 @@ out:
 static noinline int btrfs_ioctl_ino_lookup(struct file *file,
 					   void __user *argp)
 {
-	 struct btrfs_ioctl_ino_lookup_args *args;
-	 struct inode *inode;
+	struct btrfs_ioctl_ino_lookup_args *args;
+	struct inode *inode;
 	int ret = 0;
 
 	args = memdup_user(argp, sizeof(*args));
@@ -2941,8 +2927,14 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 		ret = btrfs_defrag_root(root);
 		break;
 	case S_IFREG:
-		if (!(file->f_mode & FMODE_WRITE)) {
-			ret = -EINVAL;
+		/*
+		 * Note that this does not check the file descriptor for write
+		 * access. This prevents defragmenting executables that are
+		 * running and allows defrag on files open in read-only mode.
+		 */
+		if (!capable(CAP_SYS_ADMIN) &&
+		    inode_permission(inode, MAY_WRITE)) {
+			ret = -EPERM;
 			goto out;
 		}
 
@@ -3165,10 +3157,8 @@ static long btrfs_ioctl_dev_info(struct btrfs_fs_info *fs_info,
 	di_args->total_bytes = btrfs_device_get_total_bytes(dev);
 	memcpy(di_args->uuid, dev->uuid, sizeof(di_args->uuid));
 	if (dev->name) {
-		struct rcu_string *name;
-
-		name = rcu_dereference(dev->name);
-		strncpy(di_args->path, name->str, sizeof(di_args->path) - 1);
+		strncpy(di_args->path, rcu_str_deref(dev->name),
+				sizeof(di_args->path) - 1);
 		di_args->path[sizeof(di_args->path) - 1] = 0;
 	} else {
 		di_args->path[0] = '\0';
@@ -5118,9 +5108,7 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg)
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ioctl_quota_ctl_args *sa;
-	struct btrfs_trans_handle *trans = NULL;
 	int ret;
-	int err;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -5136,28 +5124,19 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg)
 	}
 
 	down_write(&fs_info->subvol_sem);
-	trans = btrfs_start_transaction(fs_info->tree_root, 2);
-	if (IS_ERR(trans)) {
-		ret = PTR_ERR(trans);
-		goto out;
-	}
 
 	switch (sa->cmd) {
 	case BTRFS_QUOTA_CTL_ENABLE:
-		ret = btrfs_quota_enable(trans, fs_info);
+		ret = btrfs_quota_enable(fs_info);
 		break;
 	case BTRFS_QUOTA_CTL_DISABLE:
-		ret = btrfs_quota_disable(trans, fs_info);
+		ret = btrfs_quota_disable(fs_info);
 		break;
 	default:
 		ret = -EINVAL;
 		break;
 	}
 
-	err = btrfs_commit_transaction(trans);
-	if (err && !ret)
-		ret = err;
-out:
 	kfree(sa);
 	up_write(&fs_info->subvol_sem);
 drop_write:
@@ -5195,15 +5174,13 @@ static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg)
 	}
 
 	if (sa->assign) {
-		ret = btrfs_add_qgroup_relation(trans, fs_info,
-						sa->src, sa->dst);
+		ret = btrfs_add_qgroup_relation(trans, sa->src, sa->dst);
 	} else {
-		ret = btrfs_del_qgroup_relation(trans, fs_info,
-						sa->src, sa->dst);
+		ret = btrfs_del_qgroup_relation(trans, sa->src, sa->dst);
 	}
 
 	/* update qgroup status and info */
-	err = btrfs_run_qgroups(trans, fs_info);
+	err = btrfs_run_qgroups(trans);
 	if (err < 0)
 		btrfs_handle_fs_error(fs_info, err,
 				      "failed to update qgroup status and info");
@@ -5221,7 +5198,6 @@ drop_write:
 static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg)
 {
 	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_ioctl_qgroup_create_args *sa;
 	struct btrfs_trans_handle *trans;
@@ -5253,9 +5229,9 @@ static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg)
 	}
 
 	if (sa->create) {
-		ret = btrfs_create_qgroup(trans, fs_info, sa->qgroupid);
+		ret = btrfs_create_qgroup(trans, sa->qgroupid);
 	} else {
-		ret = btrfs_remove_qgroup(trans, fs_info, sa->qgroupid);
+		ret = btrfs_remove_qgroup(trans, sa->qgroupid);
 	}
 
 	err = btrfs_end_transaction(trans);
@@ -5272,7 +5248,6 @@ drop_write:
 static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg)
 {
 	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_ioctl_qgroup_limit_args *sa;
 	struct btrfs_trans_handle *trans;
@@ -5305,7 +5280,7 @@ static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg)
 		qgroupid = root->root_key.objectid;
 	}
 
-	ret = btrfs_limit_qgroup(trans, fs_info, qgroupid, &sa->lim);
+	ret = btrfs_limit_qgroup(trans, qgroupid, &sa->lim);
 
 	err = btrfs_end_transaction(trans);
 	if (err && !ret)

fs/btrfs/ordered-data.c

@@ -6,7 +6,6 @@
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/writeback.h>
-#include <linux/pagevec.h>
 #include "ctree.h"
 #include "transaction.h"
 #include "btrfs_inode.h"
@@ -421,129 +420,6 @@ out:
 	return ret == 0;
 }
 
-/* Needs to either be called under a log transaction or the log_mutex */
-void btrfs_get_logged_extents(struct btrfs_inode *inode,
-			      struct list_head *logged_list,
-			      const loff_t start,
-			      const loff_t end)
-{
-	struct btrfs_ordered_inode_tree *tree;
-	struct btrfs_ordered_extent *ordered;
-	struct rb_node *n;
-	struct rb_node *prev;
-
-	tree = &inode->ordered_tree;
-	spin_lock_irq(&tree->lock);
-	n = __tree_search(&tree->tree, end, &prev);
-	if (!n)
-		n = prev;
-	for (; n; n = rb_prev(n)) {
-		ordered = rb_entry(n, struct btrfs_ordered_extent, rb_node);
-		if (ordered->file_offset > end)
-			continue;
-		if (entry_end(ordered) <= start)
-			break;
-		if (test_and_set_bit(BTRFS_ORDERED_LOGGED, &ordered->flags))
-			continue;
-		list_add(&ordered->log_list, logged_list);
-		refcount_inc(&ordered->refs);
-	}
-	spin_unlock_irq(&tree->lock);
-}
-
-void btrfs_put_logged_extents(struct list_head *logged_list)
-{
-	struct btrfs_ordered_extent *ordered;
-
-	while (!list_empty(logged_list)) {
-		ordered = list_first_entry(logged_list,
-					   struct btrfs_ordered_extent,
-					   log_list);
-		list_del_init(&ordered->log_list);
-		btrfs_put_ordered_extent(ordered);
-	}
-}
-
-void btrfs_submit_logged_extents(struct list_head *logged_list,
-				 struct btrfs_root *log)
-{
-	int index = log->log_transid % 2;
-
-	spin_lock_irq(&log->log_extents_lock[index]);
-	list_splice_tail(logged_list, &log->logged_list[index]);
-	spin_unlock_irq(&log->log_extents_lock[index]);
-}
-
-void btrfs_wait_logged_extents(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *log, u64 transid)
-{
-	struct btrfs_ordered_extent *ordered;
-	int index = transid % 2;
-
-	spin_lock_irq(&log->log_extents_lock[index]);
-	while (!list_empty(&log->logged_list[index])) {
-		struct inode *inode;
-		ordered = list_first_entry(&log->logged_list[index],
-					   struct btrfs_ordered_extent,
-					   log_list);
-		list_del_init(&ordered->log_list);
-		inode = ordered->inode;
-		spin_unlock_irq(&log->log_extents_lock[index]);
-
-		if (!test_bit(BTRFS_ORDERED_IO_DONE, &ordered->flags) &&
-		    !test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags)) {
-			u64 start = ordered->file_offset;
-			u64 end = ordered->file_offset + ordered->len - 1;
-
-			WARN_ON(!inode);
-			filemap_fdatawrite_range(inode->i_mapping, start, end);
-		}
-		wait_event(ordered->wait, test_bit(BTRFS_ORDERED_IO_DONE,
-						   &ordered->flags));
-
-		/*
-		 * In order to keep us from losing our ordered extent
-		 * information when committing the transaction we have to make
-		 * sure that any logged extents are completed when we go to
-		 * commit the transaction.  To do this we simply increase the
-		 * current transactions pending_ordered counter and decrement it
-		 * when the ordered extent completes.
-		 */
-		if (!test_bit(BTRFS_ORDERED_COMPLETE, &ordered->flags)) {
-			struct btrfs_ordered_inode_tree *tree;
-
-			tree = &BTRFS_I(inode)->ordered_tree;
-			spin_lock_irq(&tree->lock);
-			if (!test_bit(BTRFS_ORDERED_COMPLETE, &ordered->flags)) {
-				set_bit(BTRFS_ORDERED_PENDING, &ordered->flags);
-				atomic_inc(&trans->transaction->pending_ordered);
-			}
-			spin_unlock_irq(&tree->lock);
-		}
-		btrfs_put_ordered_extent(ordered);
-		spin_lock_irq(&log->log_extents_lock[index]);
-	}
-	spin_unlock_irq(&log->log_extents_lock[index]);
-}
-
-void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid)
-{
-	struct btrfs_ordered_extent *ordered;
-	int index = transid % 2;
-
-	spin_lock_irq(&log->log_extents_lock[index]);
-	while (!list_empty(&log->logged_list[index])) {
-		ordered = list_first_entry(&log->logged_list[index],
-					   struct btrfs_ordered_extent,
-					   log_list);
-		list_del_init(&ordered->log_list);
-		spin_unlock_irq(&log->log_extents_lock[index]);
-		btrfs_put_ordered_extent(ordered);
-		spin_lock_irq(&log->log_extents_lock[index]);
-	}
-	spin_unlock_irq(&log->log_extents_lock[index]);
-}
-
 /*
  * used to drop a reference on an ordered extent.  This will free
  * the extent if the last reference is dropped
@@ -913,20 +789,6 @@ out:
 	return entry;
 }
 
-bool btrfs_have_ordered_extents_in_range(struct inode *inode,
-					 u64 file_offset,
-					 u64 len)
-{
-	struct btrfs_ordered_extent *oe;
-
-	oe = btrfs_lookup_ordered_range(BTRFS_I(inode), file_offset, len);
-	if (oe) {
-		btrfs_put_ordered_extent(oe);
-		return true;
-	}
-	return false;
-}
-
 /*
  * lookup and return any extent before 'file_offset'.  NULL is returned
  * if none is found

fs/btrfs/ordered-data.h

@@ -54,15 +54,11 @@ struct btrfs_ordered_sum {
 #define BTRFS_ORDERED_UPDATED_ISIZE 7 /* indicates whether this ordered extent
 				       * has done its due diligence in updating
 				       * the isize. */
-#define BTRFS_ORDERED_LOGGED_CSUM 8 /* We've logged the csums on this ordered
-				       ordered extent */
-#define BTRFS_ORDERED_TRUNCATED 9 /* Set when we have to truncate an extent */
+#define BTRFS_ORDERED_TRUNCATED 8 /* Set when we have to truncate an extent */
 
-#define BTRFS_ORDERED_LOGGED 10 /* Set when we've waited on this ordered extent
-				 * in the logging code. */
-#define BTRFS_ORDERED_PENDING 11 /* We are waiting for this ordered extent to
+#define BTRFS_ORDERED_PENDING 9 /* We are waiting for this ordered extent to
 				  * complete in the current transaction. */
-#define BTRFS_ORDERED_REGULAR 12 /* Regular IO for COW */
+#define BTRFS_ORDERED_REGULAR 10 /* Regular IO for COW */
 
 struct btrfs_ordered_extent {
 	/* logical offset in the file */
@@ -182,9 +178,6 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_range(
 		struct btrfs_inode *inode,
 		u64 file_offset,
 		u64 len);
-bool btrfs_have_ordered_extents_in_range(struct inode *inode,
-					 u64 file_offset,
-					 u64 len);
 int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
@@ -193,16 +186,6 @@ u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
 			       const u64 range_start, const u64 range_len);
 u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
 			      const u64 range_start, const u64 range_len);
-void btrfs_get_logged_extents(struct btrfs_inode *inode,
-			      struct list_head *logged_list,
-			      const loff_t start,
-			      const loff_t end);
-void btrfs_put_logged_extents(struct list_head *logged_list);
-void btrfs_submit_logged_extents(struct list_head *logged_list,
-				 struct btrfs_root *log);
-void btrfs_wait_logged_extents(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *log, u64 transid);
-void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid);
 int __init ordered_data_init(void);
 void __cold ordered_data_exit(void);
 

fs/btrfs/print-tree.c

@@ -52,17 +52,9 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 	u64 offset;
 	int ref_index = 0;
 
-	if (item_size < sizeof(*ei)) {
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		struct btrfs_extent_item_v0 *ei0;
-		BUG_ON(item_size != sizeof(*ei0));
-		ei0 = btrfs_item_ptr(eb, slot, struct btrfs_extent_item_v0);
-		pr_info("\t\textent refs %u\n",
-		       btrfs_extent_refs_v0(eb, ei0));
-		return;
-#else
-		BUG();
-#endif
+	if (unlikely(item_size < sizeof(*ei))) {
+		btrfs_print_v0_err(eb->fs_info);
+		btrfs_handle_fs_error(eb->fs_info, -EINVAL, NULL);
 	}
 
 	ei = btrfs_item_ptr(eb, slot, struct btrfs_extent_item);
@@ -133,20 +125,6 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 	WARN_ON(ptr > end);
 }
 
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-static void print_extent_ref_v0(struct extent_buffer *eb, int slot)
-{
-	struct btrfs_extent_ref_v0 *ref0;
-
-	ref0 = btrfs_item_ptr(eb, slot, struct btrfs_extent_ref_v0);
-	printk("\t\textent back ref root %llu gen %llu owner %llu num_refs %lu\n",
-		btrfs_ref_root_v0(eb, ref0),
-		btrfs_ref_generation_v0(eb, ref0),
-		btrfs_ref_objectid_v0(eb, ref0),
-		(unsigned long)btrfs_ref_count_v0(eb, ref0));
-}
-#endif
-
 static void print_uuid_item(struct extent_buffer *l, unsigned long offset,
 			    u32 item_size)
 {
@@ -267,8 +245,8 @@ void btrfs_print_leaf(struct extent_buffer *l)
 					    struct btrfs_file_extent_item);
 			if (btrfs_file_extent_type(l, fi) ==
 			    BTRFS_FILE_EXTENT_INLINE) {
-				pr_info("\t\tinline extent data size %u\n",
-				       btrfs_file_extent_inline_len(l, i, fi));
+				pr_info("\t\tinline extent data size %llu\n",
+				       btrfs_file_extent_ram_bytes(l, fi));
 				break;
 			}
 			pr_info("\t\textent data disk bytenr %llu nr %llu\n",
@@ -280,11 +258,8 @@ void btrfs_print_leaf(struct extent_buffer *l)
 			       btrfs_file_extent_ram_bytes(l, fi));
 			break;
 		case BTRFS_EXTENT_REF_V0_KEY:
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-			print_extent_ref_v0(l, i);
-#else
-			BUG();
-#endif
+			btrfs_print_v0_err(fs_info);
+			btrfs_handle_fs_error(fs_info, -EINVAL, NULL);
 			break;
 		case BTRFS_BLOCK_GROUP_ITEM_KEY:
 			bi = btrfs_item_ptr(l, i,

fs/btrfs/qgroup.c

@@ -530,11 +530,11 @@ void btrfs_free_qgroup_config(struct btrfs_fs_info *fs_info)
 	fs_info->qgroup_ulist = NULL;
 }
 
-static int add_qgroup_relation_item(struct btrfs_trans_handle *trans,
-				    struct btrfs_root *quota_root,
-				    u64 src, u64 dst)
+static int add_qgroup_relation_item(struct btrfs_trans_handle *trans, u64 src,
+				    u64 dst)
 {
 	int ret;
+	struct btrfs_root *quota_root = trans->fs_info->quota_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 
@@ -554,11 +554,11 @@ static int add_qgroup_relation_item(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-static int del_qgroup_relation_item(struct btrfs_trans_handle *trans,
-				    struct btrfs_root *quota_root,
-				    u64 src, u64 dst)
+static int del_qgroup_relation_item(struct btrfs_trans_handle *trans, u64 src,
+				    u64 dst)
 {
 	int ret;
+	struct btrfs_root *quota_root = trans->fs_info->quota_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 
@@ -653,10 +653,10 @@ out:
 	return ret;
 }
 
-static int del_qgroup_item(struct btrfs_trans_handle *trans,
-			   struct btrfs_root *quota_root, u64 qgroupid)
+static int del_qgroup_item(struct btrfs_trans_handle *trans, u64 qgroupid)
 {
 	int ret;
+	struct btrfs_root *quota_root = trans->fs_info->quota_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 
@@ -700,9 +700,9 @@ out:
 }
 
 static int update_qgroup_limit_item(struct btrfs_trans_handle *trans,
-				    struct btrfs_root *root,
 				    struct btrfs_qgroup *qgroup)
 {
+	struct btrfs_root *quota_root = trans->fs_info->quota_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct extent_buffer *l;
@@ -718,7 +718,7 @@ static int update_qgroup_limit_item(struct btrfs_trans_handle *trans,
 	if (!path)
 		return -ENOMEM;
 
-	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
+	ret = btrfs_search_slot(trans, quota_root, &key, path, 0, 1);
 	if (ret > 0)
 		ret = -ENOENT;
 
@@ -742,9 +742,10 @@ out:
 }
 
 static int update_qgroup_info_item(struct btrfs_trans_handle *trans,
-				   struct btrfs_root *root,
 				   struct btrfs_qgroup *qgroup)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_root *quota_root = fs_info->quota_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct extent_buffer *l;
@@ -752,7 +753,7 @@ static int update_qgroup_info_item(struct btrfs_trans_handle *trans,
 	int ret;
 	int slot;
 
-	if (btrfs_is_testing(root->fs_info))
+	if (btrfs_is_testing(fs_info))
 		return 0;
 
 	key.objectid = 0;
@@ -763,7 +764,7 @@ static int update_qgroup_info_item(struct btrfs_trans_handle *trans,
 	if (!path)
 		return -ENOMEM;
 
-	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
+	ret = btrfs_search_slot(trans, quota_root, &key, path, 0, 1);
 	if (ret > 0)
 		ret = -ENOENT;
 
@@ -786,10 +787,10 @@ out:
 	return ret;
 }
 
-static int update_qgroup_status_item(struct btrfs_trans_handle *trans,
-				     struct btrfs_fs_info *fs_info,
-				    struct btrfs_root *root)
+static int update_qgroup_status_item(struct btrfs_trans_handle *trans)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_root *quota_root = fs_info->quota_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct extent_buffer *l;
@@ -805,7 +806,7 @@ static int update_qgroup_status_item(struct btrfs_trans_handle *trans,
 	if (!path)
 		return -ENOMEM;
 
-	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
+	ret = btrfs_search_slot(trans, quota_root, &key, path, 0, 1);
 	if (ret > 0)
 		ret = -ENOENT;
 
@@ -875,8 +876,7 @@ out:
 	return ret;
 }
 
-int btrfs_quota_enable(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info)
+int btrfs_quota_enable(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_root *quota_root;
 	struct btrfs_root *tree_root = fs_info->tree_root;
@@ -886,6 +886,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 	struct btrfs_key key;
 	struct btrfs_key found_key;
 	struct btrfs_qgroup *qgroup = NULL;
+	struct btrfs_trans_handle *trans = NULL;
 	int ret = 0;
 	int slot;
 
@@ -893,9 +894,25 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 	if (fs_info->quota_root)
 		goto out;
 
+	/*
+	 * 1 for quota root item
+	 * 1 for BTRFS_QGROUP_STATUS item
+	 *
+	 * Yet we also need 2*n items for a QGROUP_INFO/QGROUP_LIMIT items
+	 * per subvolume. However those are not currently reserved since it
+	 * would be a lot of overkill.
+	 */
+	trans = btrfs_start_transaction(tree_root, 2);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		trans = NULL;
+		goto out;
+	}
+
 	fs_info->qgroup_ulist = ulist_alloc(GFP_KERNEL);
 	if (!fs_info->qgroup_ulist) {
 		ret = -ENOMEM;
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -906,12 +923,14 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 				       BTRFS_QUOTA_TREE_OBJECTID);
 	if (IS_ERR(quota_root)) {
 		ret =  PTR_ERR(quota_root);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
 	path = btrfs_alloc_path();
 	if (!path) {
 		ret = -ENOMEM;
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_root;
 	}
 
@@ -921,8 +940,10 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_insert_empty_item(trans, quota_root, path, &key,
 				      sizeof(*ptr));
-	if (ret)
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_path;
+	}
 
 	leaf = path->nodes[0];
 	ptr = btrfs_item_ptr(leaf, path->slots[0],
@@ -944,9 +965,10 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 	ret = btrfs_search_slot_for_read(tree_root, &key, path, 1, 0);
 	if (ret > 0)
 		goto out_add_root;
-	if (ret < 0)
+	if (ret < 0) {
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_path;
-
+	}
 
 	while (1) {
 		slot = path->slots[0];
@@ -956,18 +978,23 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 		if (found_key.type == BTRFS_ROOT_REF_KEY) {
 			ret = add_qgroup_item(trans, quota_root,
 					      found_key.offset);
-			if (ret)
+			if (ret) {
+				btrfs_abort_transaction(trans, ret);
 				goto out_free_path;
+			}
 
 			qgroup = add_qgroup_rb(fs_info, found_key.offset);
 			if (IS_ERR(qgroup)) {
 				ret = PTR_ERR(qgroup);
+				btrfs_abort_transaction(trans, ret);
 				goto out_free_path;
 			}
 		}
 		ret = btrfs_next_item(tree_root, path);
-		if (ret < 0)
+		if (ret < 0) {
+			btrfs_abort_transaction(trans, ret);
 			goto out_free_path;
+		}
 		if (ret)
 			break;
 	}
@@ -975,18 +1002,28 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 out_add_root:
 	btrfs_release_path(path);
 	ret = add_qgroup_item(trans, quota_root, BTRFS_FS_TREE_OBJECTID);
-	if (ret)
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_path;
+	}
 
 	qgroup = add_qgroup_rb(fs_info, BTRFS_FS_TREE_OBJECTID);
 	if (IS_ERR(qgroup)) {
 		ret = PTR_ERR(qgroup);
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_path;
 	}
 	spin_lock(&fs_info->qgroup_lock);
 	fs_info->quota_root = quota_root;
 	set_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
 	spin_unlock(&fs_info->qgroup_lock);
+
+	ret = btrfs_commit_transaction(trans);
+	if (ret) {
+		trans = NULL;
+		goto out_free_path;
+	}
+
 	ret = qgroup_rescan_init(fs_info, 0, 1);
 	if (!ret) {
 	        qgroup_rescan_zero_tracking(fs_info);
@@ -1006,20 +1043,35 @@ out:
 	if (ret) {
 		ulist_free(fs_info->qgroup_ulist);
 		fs_info->qgroup_ulist = NULL;
+		if (trans)
+			btrfs_end_transaction(trans);
 	}
 	mutex_unlock(&fs_info->qgroup_ioctl_lock);
 	return ret;
 }
 
-int btrfs_quota_disable(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info)
+int btrfs_quota_disable(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_root *quota_root;
+	struct btrfs_trans_handle *trans = NULL;
 	int ret = 0;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
 	if (!fs_info->quota_root)
 		goto out;
+
+	/*
+	 * 1 For the root item
+	 *
+	 * We should also reserve enough items for the quota tree deletion in
+	 * btrfs_clean_quota_tree but this is not done.
+	 */
+	trans = btrfs_start_transaction(fs_info->tree_root, 1);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		goto out;
+	}
+
 	clear_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
 	btrfs_qgroup_wait_for_completion(fs_info, false);
 	spin_lock(&fs_info->qgroup_lock);
@@ -1031,12 +1083,16 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
 	btrfs_free_qgroup_config(fs_info);
 
 	ret = btrfs_clean_quota_tree(trans, quota_root);
-	if (ret)
-		goto out;
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto end_trans;
+	}
 
-	ret = btrfs_del_root(trans, fs_info, &quota_root->root_key);
-	if (ret)
-		goto out;
+	ret = btrfs_del_root(trans, &quota_root->root_key);
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto end_trans;
+	}
 
 	list_del(&quota_root->dirty_list);
 
@@ -1048,6 +1104,9 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
 	free_extent_buffer(quota_root->node);
 	free_extent_buffer(quota_root->commit_root);
 	kfree(quota_root);
+
+end_trans:
+	ret = btrfs_end_transaction(trans);
 out:
 	mutex_unlock(&fs_info->qgroup_ioctl_lock);
 	return ret;
@@ -1177,9 +1236,10 @@ out:
 	return ret;
 }
 
-int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info, u64 src, u64 dst)
+int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
+			      u64 dst)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *parent;
 	struct btrfs_qgroup *member;
@@ -1216,13 +1276,13 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
 		}
 	}
 
-	ret = add_qgroup_relation_item(trans, quota_root, src, dst);
+	ret = add_qgroup_relation_item(trans, src, dst);
 	if (ret)
 		goto out;
 
-	ret = add_qgroup_relation_item(trans, quota_root, dst, src);
+	ret = add_qgroup_relation_item(trans, dst, src);
 	if (ret) {
-		del_qgroup_relation_item(trans, quota_root, src, dst);
+		del_qgroup_relation_item(trans, src, dst);
 		goto out;
 	}
 
@@ -1240,9 +1300,10 @@ out:
 	return ret;
 }
 
-static int __del_qgroup_relation(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info, u64 src, u64 dst)
+static int __del_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
+				 u64 dst)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *parent;
 	struct btrfs_qgroup *member;
@@ -1276,8 +1337,8 @@ static int __del_qgroup_relation(struct btrfs_trans_handle *trans,
 	ret = -ENOENT;
 	goto out;
 exist:
-	ret = del_qgroup_relation_item(trans, quota_root, src, dst);
-	err = del_qgroup_relation_item(trans, quota_root, dst, src);
+	ret = del_qgroup_relation_item(trans, src, dst);
+	err = del_qgroup_relation_item(trans, dst, src);
 	if (err && !ret)
 		ret = err;
 
@@ -1290,21 +1351,22 @@ out:
 	return ret;
 }
 
-int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info, u64 src, u64 dst)
+int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
+			      u64 dst)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret = 0;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	ret = __del_qgroup_relation(trans, fs_info, src, dst);
+	ret = __del_qgroup_relation(trans, src, dst);
 	mutex_unlock(&fs_info->qgroup_ioctl_lock);
 
 	return ret;
 }
 
-int btrfs_create_qgroup(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u64 qgroupid)
+int btrfs_create_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	int ret = 0;
@@ -1336,9 +1398,9 @@ out:
 	return ret;
 }
 
-int btrfs_remove_qgroup(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u64 qgroupid)
+int btrfs_remove_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	struct btrfs_qgroup_list *list;
@@ -1362,16 +1424,15 @@ int btrfs_remove_qgroup(struct btrfs_trans_handle *trans,
 			goto out;
 		}
 	}
-	ret = del_qgroup_item(trans, quota_root, qgroupid);
+	ret = del_qgroup_item(trans, qgroupid);
 	if (ret && ret != -ENOENT)
 		goto out;
 
 	while (!list_empty(&qgroup->groups)) {
 		list = list_first_entry(&qgroup->groups,
 					struct btrfs_qgroup_list, next_group);
-		ret = __del_qgroup_relation(trans, fs_info,
-					   qgroupid,
-					   list->group->qgroupid);
+		ret = __del_qgroup_relation(trans, qgroupid,
+					    list->group->qgroupid);
 		if (ret)
 			goto out;
 	}
@@ -1384,10 +1445,10 @@ out:
 	return ret;
 }
 
-int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info, u64 qgroupid,
+int btrfs_limit_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid,
 		       struct btrfs_qgroup_limit *limit)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	int ret = 0;
@@ -1451,7 +1512,7 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
 
 	spin_unlock(&fs_info->qgroup_lock);
 
-	ret = update_qgroup_limit_item(trans, quota_root, qgroup);
+	ret = update_qgroup_limit_item(trans, qgroup);
 	if (ret) {
 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
 		btrfs_info(fs_info, "unable to update quota limit for %llu",
@@ -1519,10 +1580,10 @@ int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
-		struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes,
-		gfp_t gfp_flag)
+int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans, u64 bytenr,
+			      u64 num_bytes, gfp_t gfp_flag)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_qgroup_extent_record *record;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	int ret;
@@ -1530,8 +1591,6 @@ int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)
 	    || bytenr == 0 || num_bytes == 0)
 		return 0;
-	if (WARN_ON(trans == NULL))
-		return -EINVAL;
 	record = kmalloc(sizeof(*record), gfp_flag);
 	if (!record)
 		return -ENOMEM;
@@ -1552,9 +1611,9 @@ int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
 }
 
 int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct extent_buffer *eb)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int nr = btrfs_header_nritems(eb);
 	int i, extent_type, ret;
 	struct btrfs_key key;
@@ -1584,8 +1643,8 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
 
 		num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi);
 
-		ret = btrfs_qgroup_trace_extent(trans, fs_info, bytenr,
-						num_bytes, GFP_NOFS);
+		ret = btrfs_qgroup_trace_extent(trans, bytenr, num_bytes,
+						GFP_NOFS);
 		if (ret)
 			return ret;
 	}
@@ -1655,11 +1714,10 @@ static int adjust_slots_upwards(struct btrfs_path *path, int root_level)
 }
 
 int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root,
 			       struct extent_buffer *root_eb,
 			       u64 root_gen, int root_level)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret = 0;
 	int level;
 	struct extent_buffer *eb = root_eb;
@@ -1678,7 +1736,7 @@ int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 	}
 
 	if (root_level == 0) {
-		ret = btrfs_qgroup_trace_leaf_items(trans, fs_info, root_eb);
+		ret = btrfs_qgroup_trace_leaf_items(trans, root_eb);
 		goto out;
 	}
 
@@ -1736,8 +1794,7 @@ walk_down:
 			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
 			path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
 
-			ret = btrfs_qgroup_trace_extent(trans, fs_info,
-							child_bytenr,
+			ret = btrfs_qgroup_trace_extent(trans, child_bytenr,
 							fs_info->nodesize,
 							GFP_NOFS);
 			if (ret)
@@ -1745,8 +1802,8 @@ walk_down:
 		}
 
 		if (level == 0) {
-			ret = btrfs_qgroup_trace_leaf_items(trans,fs_info,
-							   path->nodes[level]);
+			ret = btrfs_qgroup_trace_leaf_items(trans,
+							    path->nodes[level]);
 			if (ret)
 				goto out;
 
@@ -1981,12 +2038,11 @@ static int maybe_fs_roots(struct ulist *roots)
 	return is_fstree(unode->val);
 }
 
-int
-btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info,
-			    u64 bytenr, u64 num_bytes,
-			    struct ulist *old_roots, struct ulist *new_roots)
+int btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, u64 bytenr,
+				u64 num_bytes, struct ulist *old_roots,
+				struct ulist *new_roots)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct ulist *qgroups = NULL;
 	struct ulist *tmp = NULL;
 	u64 seq;
@@ -2116,9 +2172,10 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans)
 				ulist_del(record->old_roots, qgroup_to_skip,
 					  0);
 			}
-			ret = btrfs_qgroup_account_extent(trans, fs_info,
-					record->bytenr, record->num_bytes,
-					record->old_roots, new_roots);
+			ret = btrfs_qgroup_account_extent(trans, record->bytenr,
+							  record->num_bytes,
+							  record->old_roots,
+							  new_roots);
 			record->old_roots = NULL;
 			new_roots = NULL;
 		}
@@ -2136,9 +2193,9 @@ cleanup:
 /*
  * called from commit_transaction. Writes all changed qgroups to disk.
  */
-int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info)
+int btrfs_run_qgroups(struct btrfs_trans_handle *trans)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root = fs_info->quota_root;
 	int ret = 0;
 
@@ -2152,11 +2209,11 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
 					  struct btrfs_qgroup, dirty);
 		list_del_init(&qgroup->dirty);
 		spin_unlock(&fs_info->qgroup_lock);
-		ret = update_qgroup_info_item(trans, quota_root, qgroup);
+		ret = update_qgroup_info_item(trans, qgroup);
 		if (ret)
 			fs_info->qgroup_flags |=
 					BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
-		ret = update_qgroup_limit_item(trans, quota_root, qgroup);
+		ret = update_qgroup_limit_item(trans, qgroup);
 		if (ret)
 			fs_info->qgroup_flags |=
 					BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
@@ -2168,7 +2225,7 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
 		fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_ON;
 	spin_unlock(&fs_info->qgroup_lock);
 
-	ret = update_qgroup_status_item(trans, fs_info, quota_root);
+	ret = update_qgroup_status_item(trans);
 	if (ret)
 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
 
@@ -2181,13 +2238,13 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
  * cause a transaction abort so we take extra care here to only error
  * when a readonly fs is a reasonable outcome.
  */
-int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
-			 struct btrfs_fs_info *fs_info, u64 srcid, u64 objectid,
-			 struct btrfs_qgroup_inherit *inherit)
+int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
+			 u64 objectid, struct btrfs_qgroup_inherit *inherit)
 {
 	int ret = 0;
 	int i;
 	u64 *i_qgroups;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *quota_root = fs_info->quota_root;
 	struct btrfs_qgroup *srcgroup;
 	struct btrfs_qgroup *dstgroup;
@@ -2229,22 +2286,6 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto out;
 
-	if (srcid) {
-		struct btrfs_root *srcroot;
-		struct btrfs_key srckey;
-
-		srckey.objectid = srcid;
-		srckey.type = BTRFS_ROOT_ITEM_KEY;
-		srckey.offset = (u64)-1;
-		srcroot = btrfs_read_fs_root_no_name(fs_info, &srckey);
-		if (IS_ERR(srcroot)) {
-			ret = PTR_ERR(srcroot);
-			goto out;
-		}
-
-		level_size = fs_info->nodesize;
-	}
-
 	/*
 	 * add qgroup to all inherited groups
 	 */
@@ -2253,12 +2294,12 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 		for (i = 0; i < inherit->num_qgroups; ++i, ++i_qgroups) {
 			if (*i_qgroups == 0)
 				continue;
-			ret = add_qgroup_relation_item(trans, quota_root,
-						       objectid, *i_qgroups);
+			ret = add_qgroup_relation_item(trans, objectid,
+						       *i_qgroups);
 			if (ret && ret != -EEXIST)
 				goto out;
-			ret = add_qgroup_relation_item(trans, quota_root,
-						       *i_qgroups, objectid);
+			ret = add_qgroup_relation_item(trans, *i_qgroups,
+						       objectid);
 			if (ret && ret != -EEXIST)
 				goto out;
 		}
@@ -2281,7 +2322,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 		dstgroup->rsv_rfer = inherit->lim.rsv_rfer;
 		dstgroup->rsv_excl = inherit->lim.rsv_excl;
 
-		ret = update_qgroup_limit_item(trans, quota_root, dstgroup);
+		ret = update_qgroup_limit_item(trans, dstgroup);
 		if (ret) {
 			fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
 			btrfs_info(fs_info,
@@ -2301,6 +2342,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 		 * our counts don't go crazy, so at this point the only
 		 * difference between the two roots should be the root node.
 		 */
+		level_size = fs_info->nodesize;
 		dstgroup->rfer = srcgroup->rfer;
 		dstgroup->rfer_cmpr = srcgroup->rfer_cmpr;
 		dstgroup->excl = level_size;
@@ -2598,10 +2640,10 @@ static bool is_last_leaf(struct btrfs_path *path)
  * returns < 0 on error, 0 when more leafs are to be scanned.
  * returns 1 when done.
  */
-static int
-qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
-		   struct btrfs_trans_handle *trans)
+static int qgroup_rescan_leaf(struct btrfs_trans_handle *trans,
+			      struct btrfs_path *path)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_key found;
 	struct extent_buffer *scratch_leaf = NULL;
 	struct ulist *roots = NULL;
@@ -2669,8 +2711,8 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 		if (ret < 0)
 			goto out;
 		/* For rescan, just pass old_roots as NULL */
-		ret = btrfs_qgroup_account_extent(trans, fs_info,
-				found.objectid, num_bytes, NULL, roots);
+		ret = btrfs_qgroup_account_extent(trans, found.objectid,
+						  num_bytes, NULL, roots);
 		if (ret < 0)
 			goto out;
 	}
@@ -2716,7 +2758,7 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 		if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) {
 			err = -EINTR;
 		} else {
-			err = qgroup_rescan_leaf(fs_info, path, trans);
+			err = qgroup_rescan_leaf(trans, path);
 		}
 		if (err > 0)
 			btrfs_commit_transaction(trans);
@@ -2751,7 +2793,7 @@ out:
 			  err);
 		goto done;
 	}
-	ret = update_qgroup_status_item(trans, fs_info, fs_info->quota_root);
+	ret = update_qgroup_status_item(trans);
 	if (ret < 0) {
 		err = ret;
 		btrfs_err(fs_info, "fail to update qgroup status: %d", err);

fs/btrfs/qgroup.h

@@ -141,24 +141,19 @@ struct btrfs_qgroup {
 #define QGROUP_RELEASE		(1<<1)
 #define QGROUP_FREE		(1<<2)
 
-int btrfs_quota_enable(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info);
-int btrfs_quota_disable(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info);
+int btrfs_quota_enable(struct btrfs_fs_info *fs_info);
+int btrfs_quota_disable(struct btrfs_fs_info *fs_info);
 int btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info);
 void btrfs_qgroup_rescan_resume(struct btrfs_fs_info *fs_info);
 int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info,
 				     bool interruptible);
-int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info, u64 src, u64 dst);
-int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info, u64 src, u64 dst);
-int btrfs_create_qgroup(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u64 qgroupid);
-int btrfs_remove_qgroup(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info, u64 qgroupid);
-int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info, u64 qgroupid,
+int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
+			      u64 dst);
+int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
+			      u64 dst);
+int btrfs_create_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid);
+int btrfs_remove_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid);
+int btrfs_limit_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid,
 		       struct btrfs_qgroup_limit *limit);
 int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info);
 void btrfs_free_qgroup_config(struct btrfs_fs_info *fs_info);
@@ -217,9 +212,8 @@ int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info,
  * Return <0 for error, like memory allocation failure or invalid parameter
  * (NULL trans)
  */
-int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
-		struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes,
-		gfp_t gfp_flag);
+int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans, u64 bytenr,
+			      u64 num_bytes, gfp_t gfp_flag);
 
 /*
  * Inform qgroup to trace all leaf items of data
@@ -228,7 +222,6 @@ int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
  * Return <0 for error(ENOMEM)
  */
 int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct extent_buffer *eb);
 /*
  * Inform qgroup to trace a whole subtree, including all its child tree
@@ -241,20 +234,15 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
  * Return <0 for error(ENOMEM or tree search error)
  */
 int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root,
 			       struct extent_buffer *root_eb,
 			       u64 root_gen, int root_level);
-int
-btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info,
-			    u64 bytenr, u64 num_bytes,
-			    struct ulist *old_roots, struct ulist *new_roots);
+int btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, u64 bytenr,
+				u64 num_bytes, struct ulist *old_roots,
+				struct ulist *new_roots);
 int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans);
-int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info);
-int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
-			 struct btrfs_fs_info *fs_info, u64 srcid, u64 objectid,
-			 struct btrfs_qgroup_inherit *inherit);
+int btrfs_run_qgroups(struct btrfs_trans_handle *trans);
+int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
+			 u64 objectid, struct btrfs_qgroup_inherit *inherit);
 void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 			       u64 ref_root, u64 num_bytes,
 			       enum btrfs_qgroup_rsv_type type);

fs/btrfs/raid56.c

@@ -5,32 +5,19 @@
  */
 
 #include <linux/sched.h>
-#include <linux/wait.h>
 #include <linux/bio.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
 #include <linux/blkdev.h>
-#include <linux/random.h>
-#include <linux/iocontext.h>
-#include <linux/capability.h>
-#include <linux/ratelimit.h>
-#include <linux/kthread.h>
 #include <linux/raid/pq.h>
 #include <linux/hash.h>
 #include <linux/list_sort.h>
 #include <linux/raid/xor.h>
 #include <linux/mm.h>
-#include <asm/div64.h>
 #include "ctree.h"
-#include "extent_map.h"
 #include "disk-io.h"
-#include "transaction.h"
-#include "print-tree.h"
 #include "volumes.h"
 #include "raid56.h"
 #include "async-thread.h"
-#include "check-integrity.h"
-#include "rcu-string.h"
 
 /* set when additional merges to this rbio are not allowed */
 #define RBIO_RMW_LOCKED_BIT	1
@@ -175,8 +162,6 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio);
 static noinline void finish_rmw(struct btrfs_raid_bio *rbio);
 static void rmw_work(struct btrfs_work *work);
 static void read_rebuild_work(struct btrfs_work *work);
-static void async_rmw_stripe(struct btrfs_raid_bio *rbio);
-static void async_read_rebuild(struct btrfs_raid_bio *rbio);
 static int fail_bio_stripe(struct btrfs_raid_bio *rbio, struct bio *bio);
 static int fail_rbio_index(struct btrfs_raid_bio *rbio, int failed);
 static void __free_raid_bio(struct btrfs_raid_bio *rbio);
@@ -185,7 +170,13 @@ static int alloc_rbio_pages(struct btrfs_raid_bio *rbio);
 
 static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 					 int need_check);
-static void async_scrub_parity(struct btrfs_raid_bio *rbio);
+static void scrub_parity_work(struct btrfs_work *work);
+
+static void start_async_work(struct btrfs_raid_bio *rbio, btrfs_func_t work_func)
+{
+	btrfs_init_work(&rbio->work, btrfs_rmw_helper, work_func, NULL, NULL);
+	btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work);
+}
 
 /*
  * the stripe hash table is used for locking, and to collect
@@ -260,7 +251,7 @@ static void cache_rbio_pages(struct btrfs_raid_bio *rbio)
 		s = kmap(rbio->bio_pages[i]);
 		d = kmap(rbio->stripe_pages[i]);
 
-		memcpy(d, s, PAGE_SIZE);
+		copy_page(d, s);
 
 		kunmap(rbio->bio_pages[i]);
 		kunmap(rbio->stripe_pages[i]);
@@ -516,32 +507,21 @@ static void run_xor(void **pages, int src_cnt, ssize_t len)
 }
 
 /*
- * returns true if the bio list inside this rbio
- * covers an entire stripe (no rmw required).
- * Must be called with the bio list lock held, or
- * at a time when you know it is impossible to add
- * new bios into the list
+ * Returns true if the bio list inside this rbio covers an entire stripe (no
+ * rmw required).
  */
-static int __rbio_is_full(struct btrfs_raid_bio *rbio)
+static int rbio_is_full(struct btrfs_raid_bio *rbio)
 {
+	unsigned long flags;
 	unsigned long size = rbio->bio_list_bytes;
 	int ret = 1;
 
+	spin_lock_irqsave(&rbio->bio_list_lock, flags);
 	if (size != rbio->nr_data * rbio->stripe_len)
 		ret = 0;
-
 	BUG_ON(size > rbio->nr_data * rbio->stripe_len);
-	return ret;
-}
-
-static int rbio_is_full(struct btrfs_raid_bio *rbio)
-{
-	unsigned long flags;
-	int ret;
-
-	spin_lock_irqsave(&rbio->bio_list_lock, flags);
-	ret = __rbio_is_full(rbio);
 	spin_unlock_irqrestore(&rbio->bio_list_lock, flags);
+
 	return ret;
 }
 
@@ -812,16 +792,16 @@ static noinline void unlock_stripe(struct btrfs_raid_bio *rbio)
 			spin_unlock_irqrestore(&h->lock, flags);
 
 			if (next->operation == BTRFS_RBIO_READ_REBUILD)
-				async_read_rebuild(next);
+				start_async_work(next, read_rebuild_work);
 			else if (next->operation == BTRFS_RBIO_REBUILD_MISSING) {
 				steal_rbio(rbio, next);
-				async_read_rebuild(next);
+				start_async_work(next, read_rebuild_work);
 			} else if (next->operation == BTRFS_RBIO_WRITE) {
 				steal_rbio(rbio, next);
-				async_rmw_stripe(next);
+				start_async_work(next, rmw_work);
 			} else if (next->operation == BTRFS_RBIO_PARITY_SCRUB) {
 				steal_rbio(rbio, next);
-				async_scrub_parity(next);
+				start_async_work(next, scrub_parity_work);
 			}
 
 			goto done_nolock;
@@ -1275,7 +1255,7 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 						pointers);
 		} else {
 			/* raid5 */
-			memcpy(pointers[nr_data], pointers[0], PAGE_SIZE);
+			copy_page(pointers[nr_data], pointers[0]);
 			run_xor(pointers + 1, nr_data - 1, PAGE_SIZE);
 		}
 
@@ -1343,7 +1323,7 @@ write_data:
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_write_end_io;
-		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+		bio->bi_opf = REQ_OP_WRITE;
 
 		submit_bio(bio);
 	}
@@ -1508,20 +1488,6 @@ cleanup:
 	rbio_orig_end_io(rbio, BLK_STS_IOERR);
 }
 
-static void async_rmw_stripe(struct btrfs_raid_bio *rbio)
-{
-	btrfs_init_work(&rbio->work, btrfs_rmw_helper, rmw_work, NULL, NULL);
-	btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work);
-}
-
-static void async_read_rebuild(struct btrfs_raid_bio *rbio)
-{
-	btrfs_init_work(&rbio->work, btrfs_rmw_helper,
-			read_rebuild_work, NULL, NULL);
-
-	btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work);
-}
-
 /*
  * the stripe must be locked by the caller.  It will
  * unlock after all the writes are done
@@ -1599,7 +1565,7 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_rmw_end_io;
-		bio_set_op_attrs(bio, REQ_OP_READ, 0);
+		bio->bi_opf = REQ_OP_READ;
 
 		btrfs_bio_wq_end_io(rbio->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
 
@@ -1652,7 +1618,7 @@ static int partial_stripe_write(struct btrfs_raid_bio *rbio)
 
 	ret = lock_stripe_add(rbio);
 	if (ret == 0)
-		async_rmw_stripe(rbio);
+		start_async_work(rbio, rmw_work);
 	return 0;
 }
 
@@ -1720,8 +1686,11 @@ static void run_plug(struct btrfs_plug_cb *plug)
 		list_del_init(&cur->plug_list);
 
 		if (rbio_is_full(cur)) {
+			int ret;
+
 			/* we have a full stripe, send it down */
-			full_stripe_write(cur);
+			ret = full_stripe_write(cur);
+			BUG_ON(ret);
 			continue;
 		}
 		if (last) {
@@ -1941,9 +1910,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 			BUG_ON(failb != -1);
 pstripe:
 			/* Copy parity block into failed block to start with */
-			memcpy(pointers[faila],
-			       pointers[rbio->nr_data],
-			       PAGE_SIZE);
+			copy_page(pointers[faila], pointers[rbio->nr_data]);
 
 			/* rearrange the pointer array */
 			p = pointers[faila];
@@ -2145,7 +2112,7 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_recover_end_io;
-		bio_set_op_attrs(bio, REQ_OP_READ, 0);
+		bio->bi_opf = REQ_OP_READ;
 
 		btrfs_bio_wq_end_io(rbio->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
 
@@ -2448,7 +2415,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 						pointers);
 		} else {
 			/* raid5 */
-			memcpy(pointers[nr_data], pointers[0], PAGE_SIZE);
+			copy_page(pointers[nr_data], pointers[0]);
 			run_xor(pointers + 1, nr_data - 1, PAGE_SIZE);
 		}
 
@@ -2456,7 +2423,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 		p = rbio_stripe_page(rbio, rbio->scrubp, pagenr);
 		parity = kmap(p);
 		if (memcmp(parity, pointers[rbio->scrubp], PAGE_SIZE))
-			memcpy(parity, pointers[rbio->scrubp], PAGE_SIZE);
+			copy_page(parity, pointers[rbio->scrubp]);
 		else
 			/* Parity is right, needn't writeback */
 			bitmap_clear(rbio->dbitmap, pagenr, 1);
@@ -2517,7 +2484,7 @@ submit_write:
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_write_end_io;
-		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+		bio->bi_opf = REQ_OP_WRITE;
 
 		submit_bio(bio);
 	}
@@ -2699,7 +2666,7 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio)
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid56_parity_scrub_end_io;
-		bio_set_op_attrs(bio, REQ_OP_READ, 0);
+		bio->bi_opf = REQ_OP_READ;
 
 		btrfs_bio_wq_end_io(rbio->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
 
@@ -2728,18 +2695,10 @@ static void scrub_parity_work(struct btrfs_work *work)
 	raid56_parity_scrub_stripe(rbio);
 }
 
-static void async_scrub_parity(struct btrfs_raid_bio *rbio)
-{
-	btrfs_init_work(&rbio->work, btrfs_rmw_helper,
-			scrub_parity_work, NULL, NULL);
-
-	btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work);
-}
-
 void raid56_parity_submit_scrub_rbio(struct btrfs_raid_bio *rbio)
 {
 	if (!lock_stripe_add(rbio))
-		async_scrub_parity(rbio);
+		start_async_work(rbio, scrub_parity_work);
 }
 
 /* The following code is used for dev replace of a missing RAID 5/6 device. */
@@ -2781,5 +2740,5 @@ raid56_alloc_missing_rbio(struct btrfs_fs_info *fs_info, struct bio *bio,
 void raid56_submit_missing_rbio(struct btrfs_raid_bio *rbio)
 {
 	if (!lock_stripe_add(rbio))
-		async_read_rebuild(rbio);
+		start_async_work(rbio, read_rebuild_work);
 }

fs/btrfs/reada.c

@@ -7,7 +7,6 @@
 #include <linux/pagemap.h>
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
-#include <linux/rbtree.h>
 #include <linux/slab.h>
 #include <linux/workqueue.h>
 #include "ctree.h"
@@ -355,7 +354,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
 		dev = bbio->stripes[nzones].dev;
 
 		/* cannot read ahead on missing device. */
-		 if (!dev->bdev)
+		if (!dev->bdev)
 			continue;
 
 		zone = reada_find_zone(dev, logical, bbio);

fs/btrfs/relocation.c

@@ -586,29 +586,6 @@ static struct btrfs_root *read_fs_root(struct btrfs_fs_info *fs_info,
 	return btrfs_get_fs_root(fs_info, &key, false);
 }
 
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-static noinline_for_stack
-struct btrfs_root *find_tree_root(struct reloc_control *rc,
-				  struct extent_buffer *leaf,
-				  struct btrfs_extent_ref_v0 *ref0)
-{
-	struct btrfs_root *root;
-	u64 root_objectid = btrfs_ref_root_v0(leaf, ref0);
-	u64 generation = btrfs_ref_generation_v0(leaf, ref0);
-
-	BUG_ON(root_objectid == BTRFS_TREE_RELOC_OBJECTID);
-
-	root = read_fs_root(rc->extent_root->fs_info, root_objectid);
-	BUG_ON(IS_ERR(root));
-
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
-	    generation != btrfs_root_generation(&root->root_item))
-		return NULL;
-
-	return root;
-}
-#endif
-
 static noinline_for_stack
 int find_inline_backref(struct extent_buffer *leaf, int slot,
 			unsigned long *ptr, unsigned long *end)
@@ -621,12 +598,11 @@ int find_inline_backref(struct extent_buffer *leaf, int slot,
 	btrfs_item_key_to_cpu(leaf, &key, slot);
 
 	item_size = btrfs_item_size_nr(leaf, slot);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
 	if (item_size < sizeof(*ei)) {
-		WARN_ON(item_size != sizeof(struct btrfs_extent_item_v0));
+		btrfs_print_v0_err(leaf->fs_info);
+		btrfs_handle_fs_error(leaf->fs_info, -EINVAL, NULL);
 		return 1;
 	}
-#endif
 	ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
 	WARN_ON(!(btrfs_extent_flags(leaf, ei) &
 		  BTRFS_EXTENT_FLAG_TREE_BLOCK));
@@ -792,7 +768,7 @@ again:
 			type = btrfs_get_extent_inline_ref_type(eb, iref,
 							BTRFS_REF_TYPE_BLOCK);
 			if (type == BTRFS_REF_TYPE_INVALID) {
-				err = -EINVAL;
+				err = -EUCLEAN;
 				goto out;
 			}
 			key.type = type;
@@ -811,29 +787,7 @@ again:
 			goto next;
 		}
 
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		if (key.type == BTRFS_SHARED_BLOCK_REF_KEY ||
-		    key.type == BTRFS_EXTENT_REF_V0_KEY) {
-			if (key.type == BTRFS_EXTENT_REF_V0_KEY) {
-				struct btrfs_extent_ref_v0 *ref0;
-				ref0 = btrfs_item_ptr(eb, path1->slots[0],
-						struct btrfs_extent_ref_v0);
-				if (key.objectid == key.offset) {
-					root = find_tree_root(rc, eb, ref0);
-					if (root && !should_ignore_root(root))
-						cur->root = root;
-					else
-						list_add(&cur->list, &useless);
-					break;
-				}
-				if (is_cowonly_root(btrfs_ref_root_v0(eb,
-								      ref0)))
-					cur->cowonly = 1;
-			}
-#else
-		ASSERT(key.type != BTRFS_EXTENT_REF_V0_KEY);
 		if (key.type == BTRFS_SHARED_BLOCK_REF_KEY) {
-#endif
 			if (key.objectid == key.offset) {
 				/*
 				 * only root blocks of reloc trees use
@@ -876,6 +830,12 @@ again:
 			edge->node[UPPER] = upper;
 
 			goto next;
+		} else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) {
+			err = -EINVAL;
+			btrfs_print_v0_err(rc->extent_root->fs_info);
+			btrfs_handle_fs_error(rc->extent_root->fs_info, err,
+					      NULL);
+			goto out;
 		} else if (key.type != BTRFS_TREE_BLOCK_REF_KEY) {
 			goto next;
 		}
@@ -1321,18 +1281,19 @@ static void __del_reloc_root(struct btrfs_root *root)
 	struct mapping_node *node = NULL;
 	struct reloc_control *rc = fs_info->reloc_ctl;
 
-	spin_lock(&rc->reloc_root_tree.lock);
-	rb_node = tree_search(&rc->reloc_root_tree.rb_root,
-			      root->node->start);
-	if (rb_node) {
-		node = rb_entry(rb_node, struct mapping_node, rb_node);
-		rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
+	if (rc) {
+		spin_lock(&rc->reloc_root_tree.lock);
+		rb_node = tree_search(&rc->reloc_root_tree.rb_root,
+				      root->node->start);
+		if (rb_node) {
+			node = rb_entry(rb_node, struct mapping_node, rb_node);
+			rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
+		}
+		spin_unlock(&rc->reloc_root_tree.lock);
+		if (!node)
+			return;
+		BUG_ON((struct btrfs_root *)node->data != root);
 	}
-	spin_unlock(&rc->reloc_root_tree.lock);
-
-	if (!node)
-		return;
-	BUG_ON((struct btrfs_root *)node->data != root);
 
 	spin_lock(&fs_info->trans_lock);
 	list_del_init(&root->root_list);
@@ -1918,13 +1879,12 @@ again:
 		 *    and tree block numbers, if current trans doesn't free
 		 *    data reloc tree inode.
 		 */
-		ret = btrfs_qgroup_trace_subtree(trans, src, parent,
+		ret = btrfs_qgroup_trace_subtree(trans, parent,
 				btrfs_header_generation(parent),
 				btrfs_header_level(parent));
 		if (ret < 0)
 			break;
-		ret = btrfs_qgroup_trace_subtree(trans, dest,
-				path->nodes[level],
+		ret = btrfs_qgroup_trace_subtree(trans, path->nodes[level],
 				btrfs_header_generation(path->nodes[level]),
 				btrfs_header_level(path->nodes[level]));
 		if (ret < 0)
@@ -3333,48 +3293,6 @@ int relocate_data_extent(struct inode *inode, struct btrfs_key *extent_key,
 	return 0;
 }
 
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-static int get_ref_objectid_v0(struct reloc_control *rc,
-			       struct btrfs_path *path,
-			       struct btrfs_key *extent_key,
-			       u64 *ref_objectid, int *path_change)
-{
-	struct btrfs_key key;
-	struct extent_buffer *leaf;
-	struct btrfs_extent_ref_v0 *ref0;
-	int ret;
-	int slot;
-
-	leaf = path->nodes[0];
-	slot = path->slots[0];
-	while (1) {
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(rc->extent_root, path);
-			if (ret < 0)
-				return ret;
-			BUG_ON(ret > 0);
-			leaf = path->nodes[0];
-			slot = path->slots[0];
-			if (path_change)
-				*path_change = 1;
-		}
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-		if (key.objectid != extent_key->objectid)
-			return -ENOENT;
-
-		if (key.type != BTRFS_EXTENT_REF_V0_KEY) {
-			slot++;
-			continue;
-		}
-		ref0 = btrfs_item_ptr(leaf, slot,
-				struct btrfs_extent_ref_v0);
-		*ref_objectid = btrfs_ref_objectid_v0(leaf, ref0);
-		break;
-	}
-	return 0;
-}
-#endif
-
 /*
  * helper to add a tree block to the list.
  * the major work is getting the generation and level of the block
@@ -3407,23 +3325,12 @@ static int add_tree_block(struct reloc_control *rc,
 			level = (int)extent_key->offset;
 		}
 		generation = btrfs_extent_generation(eb, ei);
+	} else if (unlikely(item_size == sizeof(struct btrfs_extent_item_v0))) {
+		btrfs_print_v0_err(eb->fs_info);
+		btrfs_handle_fs_error(eb->fs_info, -EINVAL, NULL);
+		return -EINVAL;
 	} else {
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		u64 ref_owner;
-		int ret;
-
-		BUG_ON(item_size != sizeof(struct btrfs_extent_item_v0));
-		ret = get_ref_objectid_v0(rc, path, extent_key,
-					  &ref_owner, NULL);
-		if (ret < 0)
-			return ret;
-		BUG_ON(ref_owner >= BTRFS_MAX_LEVEL);
-		level = (int)ref_owner;
-		/* FIXME: get real generation */
-		generation = 0;
-#else
 		BUG();
-#endif
 	}
 
 	btrfs_release_path(path);
@@ -3563,11 +3470,8 @@ static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
 	key.offset = 0;
 
 	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
-	if (IS_ERR(inode) || is_bad_inode(inode)) {
-		if (!IS_ERR(inode))
-			iput(inode);
+	if (IS_ERR(inode))
 		return -ENOENT;
-	}
 
 truncate:
 	ret = btrfs_check_trunc_cache_free_space(fs_info,
@@ -3781,12 +3685,7 @@ int add_data_references(struct reloc_control *rc,
 	eb = path->nodes[0];
 	ptr = btrfs_item_ptr_offset(eb, path->slots[0]);
 	end = ptr + btrfs_item_size_nr(eb, path->slots[0]);
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-	if (ptr + sizeof(struct btrfs_extent_item_v0) == end)
-		ptr = end;
-	else
-#endif
-		ptr += sizeof(struct btrfs_extent_item);
+	ptr += sizeof(struct btrfs_extent_item);
 
 	while (ptr < end) {
 		iref = (struct btrfs_extent_inline_ref *)ptr;
@@ -3801,7 +3700,7 @@ int add_data_references(struct reloc_control *rc,
 			ret = find_data_references(rc, extent_key,
 						   eb, dref, blocks);
 		} else {
-			ret = -EINVAL;
+			ret = -EUCLEAN;
 			btrfs_err(rc->extent_root->fs_info,
 		     "extent %llu slot %d has an invalid inline ref type",
 			     eb->start, path->slots[0]);
@@ -3832,13 +3731,7 @@ int add_data_references(struct reloc_control *rc,
 		if (key.objectid != extent_key->objectid)
 			break;
 
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-		if (key.type == BTRFS_SHARED_DATA_REF_KEY ||
-		    key.type == BTRFS_EXTENT_REF_V0_KEY) {
-#else
-		BUG_ON(key.type == BTRFS_EXTENT_REF_V0_KEY);
 		if (key.type == BTRFS_SHARED_DATA_REF_KEY) {
-#endif
 			ret = __add_tree_block(rc, key.offset, blocksize,
 					       blocks);
 		} else if (key.type == BTRFS_EXTENT_DATA_REF_KEY) {
@@ -3846,6 +3739,10 @@ int add_data_references(struct reloc_control *rc,
 					      struct btrfs_extent_data_ref);
 			ret = find_data_references(rc, extent_key,
 						   eb, dref, blocks);
+		} else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) {
+			btrfs_print_v0_err(eb->fs_info);
+			btrfs_handle_fs_error(eb->fs_info, -EINVAL, NULL);
+			ret = -EINVAL;
 		} else {
 			ret = 0;
 		}
@@ -4084,41 +3981,13 @@ restart:
 			flags = btrfs_extent_flags(path->nodes[0], ei);
 			ret = check_extent_flags(flags);
 			BUG_ON(ret);
-
+		} else if (unlikely(item_size == sizeof(struct btrfs_extent_item_v0))) {
+			err = -EINVAL;
+			btrfs_print_v0_err(trans->fs_info);
+			btrfs_abort_transaction(trans, err);
+			break;
 		} else {
-#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
-			u64 ref_owner;
-			int path_change = 0;
-
-			BUG_ON(item_size !=
-			       sizeof(struct btrfs_extent_item_v0));
-			ret = get_ref_objectid_v0(rc, path, &key, &ref_owner,
-						  &path_change);
-			if (ret < 0) {
-				err = ret;
-				break;
-			}
-			if (ref_owner < BTRFS_FIRST_FREE_OBJECTID)
-				flags = BTRFS_EXTENT_FLAG_TREE_BLOCK;
-			else
-				flags = BTRFS_EXTENT_FLAG_DATA;
-
-			if (path_change) {
-				btrfs_release_path(path);
-
-				path->search_commit_root = 1;
-				path->skip_locking = 1;
-				ret = btrfs_search_slot(NULL, rc->extent_root,
-							&key, path, 0, 0);
-				if (ret < 0) {
-					err = ret;
-					break;
-				}
-				BUG_ON(ret > 0);
-			}
-#else
 			BUG();
-#endif
 		}
 
 		if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
@@ -4169,8 +4038,7 @@ restart:
 		}
 	}
 	if (trans && progress && err == -ENOSPC) {
-		ret = btrfs_force_chunk_alloc(trans, fs_info,
-					      rc->block_group->flags);
+		ret = btrfs_force_chunk_alloc(trans, rc->block_group->flags);
 		if (ret == 1) {
 			err = 0;
 			progress = 0;
@@ -4284,7 +4152,7 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
-	BUG_ON(IS_ERR(inode) || is_bad_inode(inode));
+	BUG_ON(IS_ERR(inode));
 	BTRFS_I(inode)->index_cnt = group->key.objectid;
 
 	err = btrfs_orphan_add(trans, BTRFS_I(inode));
@@ -4375,7 +4243,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 	rc->block_group = btrfs_lookup_block_group(fs_info, group_start);
 	BUG_ON(!rc->block_group);
 
-	ret = btrfs_inc_block_group_ro(fs_info, rc->block_group);
+	ret = btrfs_inc_block_group_ro(rc->block_group);
 	if (ret) {
 		err = ret;
 		goto out;

fs/btrfs/root-tree.c

@@ -320,9 +320,9 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 
 /* drop the root item for 'key' from the tree root */
 int btrfs_del_root(struct btrfs_trans_handle *trans,
-		   struct btrfs_fs_info *fs_info, const struct btrfs_key *key)
+		   const struct btrfs_key *key)
 {
-	struct btrfs_root *root = fs_info->tree_root;
+	struct btrfs_root *root = trans->fs_info->tree_root;
 	struct btrfs_path *path;
 	int ret;
 
@@ -341,13 +341,12 @@ out:
 	return ret;
 }
 
-int btrfs_del_root_ref(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info,
-		       u64 root_id, u64 ref_id, u64 dirid, u64 *sequence,
-		       const char *name, int name_len)
+int btrfs_del_root_ref(struct btrfs_trans_handle *trans, u64 root_id,
+		       u64 ref_id, u64 dirid, u64 *sequence, const char *name,
+		       int name_len)
 
 {
-	struct btrfs_root *tree_root = fs_info->tree_root;
+	struct btrfs_root *tree_root = trans->fs_info->tree_root;
 	struct btrfs_path *path;
 	struct btrfs_root_ref *ref;
 	struct extent_buffer *leaf;
@@ -413,12 +412,11 @@ out:
  *
  * Will return 0, -ENOMEM, or anything from the CoW path
  */
-int btrfs_add_root_ref(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info,
-		       u64 root_id, u64 ref_id, u64 dirid, u64 sequence,
-		       const char *name, int name_len)
+int btrfs_add_root_ref(struct btrfs_trans_handle *trans, u64 root_id,
+		       u64 ref_id, u64 dirid, u64 sequence, const char *name,
+		       int name_len)
 {
-	struct btrfs_root *tree_root = fs_info->tree_root;
+	struct btrfs_root *tree_root = trans->fs_info->tree_root;
 	struct btrfs_key key;
 	int ret;
 	struct btrfs_path *path;

fs/btrfs/scrub.c

@@ -188,32 +188,6 @@ struct scrub_ctx {
 	refcount_t              refs;
 };
 
-struct scrub_fixup_nodatasum {
-	struct scrub_ctx	*sctx;
-	struct btrfs_device	*dev;
-	u64			logical;
-	struct btrfs_root	*root;
-	struct btrfs_work	work;
-	int			mirror_num;
-};
-
-struct scrub_nocow_inode {
-	u64			inum;
-	u64			offset;
-	u64			root;
-	struct list_head	list;
-};
-
-struct scrub_copy_nocow_ctx {
-	struct scrub_ctx	*sctx;
-	u64			logical;
-	u64			len;
-	int			mirror_num;
-	u64			physical_for_dev_replace;
-	struct list_head	inodes;
-	struct btrfs_work	work;
-};
-
 struct scrub_warning {
 	struct btrfs_path	*path;
 	u64			extent_item_size;
@@ -232,8 +206,6 @@ struct full_stripe_lock {
 
 static void scrub_pending_bio_inc(struct scrub_ctx *sctx);
 static void scrub_pending_bio_dec(struct scrub_ctx *sctx);
-static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx);
-static void scrub_pending_trans_workers_dec(struct scrub_ctx *sctx);
 static int scrub_handle_errored_block(struct scrub_block *sblock_to_check);
 static int scrub_setup_recheck_block(struct scrub_block *original_sblock,
 				     struct scrub_block *sblocks_for_recheck);
@@ -277,13 +249,6 @@ static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
 static void scrub_wr_submit(struct scrub_ctx *sctx);
 static void scrub_wr_bio_end_io(struct bio *bio);
 static void scrub_wr_bio_end_io_worker(struct btrfs_work *work);
-static int write_page_nocow(struct scrub_ctx *sctx,
-			    u64 physical_for_dev_replace, struct page *page);
-static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
-				      struct scrub_copy_nocow_ctx *ctx);
-static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
-			    int mirror_num, u64 physical_for_dev_replace);
-static void copy_nocow_pages_worker(struct btrfs_work *work);
 static void __scrub_blocked_if_needed(struct btrfs_fs_info *fs_info);
 static void scrub_blocked_if_needed(struct btrfs_fs_info *fs_info);
 static void scrub_put_ctx(struct scrub_ctx *sctx);
@@ -555,60 +520,6 @@ out:
 	return ret;
 }
 
-/*
- * used for workers that require transaction commits (i.e., for the
- * NOCOW case)
- */
-static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx)
-{
-	struct btrfs_fs_info *fs_info = sctx->fs_info;
-
-	refcount_inc(&sctx->refs);
-	/*
-	 * increment scrubs_running to prevent cancel requests from
-	 * completing as long as a worker is running. we must also
-	 * increment scrubs_paused to prevent deadlocking on pause
-	 * requests used for transactions commits (as the worker uses a
-	 * transaction context). it is safe to regard the worker
-	 * as paused for all matters practical. effectively, we only
-	 * avoid cancellation requests from completing.
-	 */
-	mutex_lock(&fs_info->scrub_lock);
-	atomic_inc(&fs_info->scrubs_running);
-	atomic_inc(&fs_info->scrubs_paused);
-	mutex_unlock(&fs_info->scrub_lock);
-
-	/*
-	 * check if @scrubs_running=@scrubs_paused condition
-	 * inside wait_event() is not an atomic operation.
-	 * which means we may inc/dec @scrub_running/paused
-	 * at any time. Let's wake up @scrub_pause_wait as
-	 * much as we can to let commit transaction blocked less.
-	 */
-	wake_up(&fs_info->scrub_pause_wait);
-
-	atomic_inc(&sctx->workers_pending);
-}
-
-/* used for workers that require transaction commits */
-static void scrub_pending_trans_workers_dec(struct scrub_ctx *sctx)
-{
-	struct btrfs_fs_info *fs_info = sctx->fs_info;
-
-	/*
-	 * see scrub_pending_trans_workers_inc() why we're pretending
-	 * to be paused in the scrub counters
-	 */
-	mutex_lock(&fs_info->scrub_lock);
-	atomic_dec(&fs_info->scrubs_running);
-	atomic_dec(&fs_info->scrubs_paused);
-	mutex_unlock(&fs_info->scrub_lock);
-	atomic_dec(&sctx->workers_pending);
-	wake_up(&fs_info->scrub_pause_wait);
-	wake_up(&sctx->list_wait);
-	scrub_put_ctx(sctx);
-}
-
 static void scrub_free_csums(struct scrub_ctx *sctx)
 {
 	while (!list_empty(&sctx->csum_list)) {
@@ -882,194 +793,6 @@ out:
 	btrfs_free_path(path);
 }
 
-static int scrub_fixup_readpage(u64 inum, u64 offset, u64 root, void *fixup_ctx)
-{
-	struct page *page = NULL;
-	unsigned long index;
-	struct scrub_fixup_nodatasum *fixup = fixup_ctx;
-	int ret;
-	int corrected = 0;
-	struct btrfs_key key;
-	struct inode *inode = NULL;
-	struct btrfs_fs_info *fs_info;
-	u64 end = offset + PAGE_SIZE - 1;
-	struct btrfs_root *local_root;
-	int srcu_index;
-
-	key.objectid = root;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	fs_info = fixup->root->fs_info;
-	srcu_index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	local_root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(local_root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
-		return PTR_ERR(local_root);
-	}
-
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.objectid = inum;
-	key.offset = 0;
-	inode = btrfs_iget(fs_info->sb, &key, local_root, NULL);
-	srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
-	if (IS_ERR(inode))
-		return PTR_ERR(inode);
-
-	index = offset >> PAGE_SHIFT;
-
-	page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
-	if (!page) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	if (PageUptodate(page)) {
-		if (PageDirty(page)) {
-			/*
-			 * we need to write the data to the defect sector. the
-			 * data that was in that sector is not in memory,
-			 * because the page was modified. we must not write the
-			 * modified page to that sector.
-			 *
-			 * TODO: what could be done here: wait for the delalloc
-			 *       runner to write out that page (might involve
-			 *       COW) and see whether the sector is still
-			 *       referenced afterwards.
-			 *
-			 * For the meantime, we'll treat this error
-			 * incorrectable, although there is a chance that a
-			 * later scrub will find the bad sector again and that
-			 * there's no dirty page in memory, then.
-			 */
-			ret = -EIO;
-			goto out;
-		}
-		ret = repair_io_failure(fs_info, inum, offset, PAGE_SIZE,
-					fixup->logical, page,
-					offset - page_offset(page),
-					fixup->mirror_num);
-		unlock_page(page);
-		corrected = !ret;
-	} else {
-		/*
-		 * we need to get good data first. the general readpage path
-		 * will call repair_io_failure for us, we just have to make
-		 * sure we read the bad mirror.
-		 */
-		ret = set_extent_bits(&BTRFS_I(inode)->io_tree, offset, end,
-					EXTENT_DAMAGED);
-		if (ret) {
-			/* set_extent_bits should give proper error */
-			WARN_ON(ret > 0);
-			if (ret > 0)
-				ret = -EFAULT;
-			goto out;
-		}
-
-		ret = extent_read_full_page(&BTRFS_I(inode)->io_tree, page,
-						btrfs_get_extent,
-						fixup->mirror_num);
-		wait_on_page_locked(page);
-
-		corrected = !test_range_bit(&BTRFS_I(inode)->io_tree, offset,
-						end, EXTENT_DAMAGED, 0, NULL);
-		if (!corrected)
-			clear_extent_bits(&BTRFS_I(inode)->io_tree, offset, end,
-						EXTENT_DAMAGED);
-	}
-
-out:
-	if (page)
-		put_page(page);
-
-	iput(inode);
-
-	if (ret < 0)
-		return ret;
-
-	if (ret == 0 && corrected) {
-		/*
-		 * we only need to call readpage for one of the inodes belonging
-		 * to this extent. so make iterate_extent_inodes stop
-		 */
-		return 1;
-	}
-
-	return -EIO;
-}
-
-static void scrub_fixup_nodatasum(struct btrfs_work *work)
-{
-	struct btrfs_fs_info *fs_info;
-	int ret;
-	struct scrub_fixup_nodatasum *fixup;
-	struct scrub_ctx *sctx;
-	struct btrfs_trans_handle *trans = NULL;
-	struct btrfs_path *path;
-	int uncorrectable = 0;
-
-	fixup = container_of(work, struct scrub_fixup_nodatasum, work);
-	sctx = fixup->sctx;
-	fs_info = fixup->root->fs_info;
-
-	path = btrfs_alloc_path();
-	if (!path) {
-		spin_lock(&sctx->stat_lock);
-		++sctx->stat.malloc_errors;
-		spin_unlock(&sctx->stat_lock);
-		uncorrectable = 1;
-		goto out;
-	}
-
-	trans = btrfs_join_transaction(fixup->root);
-	if (IS_ERR(trans)) {
-		uncorrectable = 1;
-		goto out;
-	}
-
-	/*
-	 * the idea is to trigger a regular read through the standard path. we
-	 * read a page from the (failed) logical address by specifying the
-	 * corresponding copynum of the failed sector. thus, that readpage is
-	 * expected to fail.
-	 * that is the point where on-the-fly error correction will kick in
-	 * (once it's finished) and rewrite the failed sector if a good copy
-	 * can be found.
-	 */
-	ret = iterate_inodes_from_logical(fixup->logical, fs_info, path,
-					  scrub_fixup_readpage, fixup, false);
-	if (ret < 0) {
-		uncorrectable = 1;
-		goto out;
-	}
-	WARN_ON(ret != 1);
-
-	spin_lock(&sctx->stat_lock);
-	++sctx->stat.corrected_errors;
-	spin_unlock(&sctx->stat_lock);
-
-out:
-	if (trans && !IS_ERR(trans))
-		btrfs_end_transaction(trans);
-	if (uncorrectable) {
-		spin_lock(&sctx->stat_lock);
-		++sctx->stat.uncorrectable_errors;
-		spin_unlock(&sctx->stat_lock);
-		btrfs_dev_replace_stats_inc(
-			&fs_info->dev_replace.num_uncorrectable_read_errors);
-		btrfs_err_rl_in_rcu(fs_info,
-		    "unable to fixup (nodatasum) error at logical %llu on dev %s",
-			fixup->logical, rcu_str_deref(fixup->dev->name));
-	}
-
-	btrfs_free_path(path);
-	kfree(fixup);
-
-	scrub_pending_trans_workers_dec(sctx);
-}
-
 static inline void scrub_get_recover(struct scrub_recover *recover)
 {
 	refcount_inc(&recover->refs);
@@ -1263,42 +986,6 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 		goto out;
 	}
 
-	/*
-	 * NOTE: Even for nodatasum case, it's still possible that it's a
-	 * compressed data extent, thus scrub_fixup_nodatasum(), which write
-	 * inode page cache onto disk, could cause serious data corruption.
-	 *
-	 * So here we could only read from disk, and hope our recovery could
-	 * reach disk before the newer write.
-	 */
-	if (0 && !is_metadata && !have_csum) {
-		struct scrub_fixup_nodatasum *fixup_nodatasum;
-
-		WARN_ON(sctx->is_dev_replace);
-
-		/*
-		 * !is_metadata and !have_csum, this means that the data
-		 * might not be COWed, that it might be modified
-		 * concurrently. The general strategy to work on the
-		 * commit root does not help in the case when COW is not
-		 * used.
-		 */
-		fixup_nodatasum = kzalloc(sizeof(*fixup_nodatasum), GFP_NOFS);
-		if (!fixup_nodatasum)
-			goto did_not_correct_error;
-		fixup_nodatasum->sctx = sctx;
-		fixup_nodatasum->dev = dev;
-		fixup_nodatasum->logical = logical;
-		fixup_nodatasum->root = fs_info->extent_root;
-		fixup_nodatasum->mirror_num = failed_mirror_index + 1;
-		scrub_pending_trans_workers_inc(sctx);
-		btrfs_init_work(&fixup_nodatasum->work, btrfs_scrub_helper,
-				scrub_fixup_nodatasum, NULL, NULL);
-		btrfs_queue_work(fs_info->scrub_workers,
-				 &fixup_nodatasum->work);
-		goto out;
-	}
-
 	/*
 	 * now build and submit the bios for the other mirrors, check
 	 * checksums.
@@ -1866,7 +1553,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 		bio = btrfs_io_bio_alloc(1);
 		bio_set_dev(bio, page_bad->dev->bdev);
 		bio->bi_iter.bi_sector = page_bad->physical >> 9;
-		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+		bio->bi_opf = REQ_OP_WRITE;
 
 		ret = bio_add_page(bio, page_good->page, PAGE_SIZE, 0);
 		if (PAGE_SIZE != ret) {
@@ -1961,7 +1648,7 @@ again:
 		bio->bi_end_io = scrub_wr_bio_end_io;
 		bio_set_dev(bio, sbio->dev->bdev);
 		bio->bi_iter.bi_sector = sbio->physical >> 9;
-		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+		bio->bi_opf = REQ_OP_WRITE;
 		sbio->status = 0;
 	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
 		   spage->physical_for_dev_replace ||
@@ -2361,7 +2048,7 @@ again:
 		bio->bi_end_io = scrub_bio_end_io;
 		bio_set_dev(bio, sbio->dev->bdev);
 		bio->bi_iter.bi_sector = sbio->physical >> 9;
-		bio_set_op_attrs(bio, REQ_OP_READ, 0);
+		bio->bi_opf = REQ_OP_READ;
 		sbio->status = 0;
 	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
 		   spage->physical ||
@@ -2800,17 +2487,10 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map,
 			have_csum = scrub_find_csum(sctx, logical, csum);
 			if (have_csum == 0)
 				++sctx->stat.no_csum;
-			if (0 && sctx->is_dev_replace && !have_csum) {
-				ret = copy_nocow_pages(sctx, logical, l,
-						       mirror_num,
-						      physical_for_dev_replace);
-				goto behind_scrub_pages;
-			}
 		}
 		ret = scrub_pages(sctx, logical, l, physical, dev, flags, gen,
 				  mirror_num, have_csum ? csum : NULL, 0,
 				  physical_for_dev_replace);
-behind_scrub_pages:
 		if (ret)
 			return ret;
 		len -= l;
@@ -3863,7 +3543,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 * -> btrfs_scrub_pause()
 		 */
 		scrub_pause_on(fs_info);
-		ret = btrfs_inc_block_group_ro(fs_info, cache);
+		ret = btrfs_inc_block_group_ro(cache);
 		if (!ret && is_dev_replace) {
 			/*
 			 * If we are doing a device replace wait for any tasks
@@ -3982,14 +3662,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		if (!cache->removed && !cache->ro && cache->reserved == 0 &&
 		    btrfs_block_group_used(&cache->item) == 0) {
 			spin_unlock(&cache->lock);
-			spin_lock(&fs_info->unused_bgs_lock);
-			if (list_empty(&cache->bg_list)) {
-				btrfs_get_block_group(cache);
-				trace_btrfs_add_unused_block_group(cache);
-				list_add_tail(&cache->bg_list,
-					      &fs_info->unused_bgs);
-			}
-			spin_unlock(&fs_info->unused_bgs_lock);
+			btrfs_mark_bg_unused(cache);
 		} else {
 			spin_unlock(&cache->lock);
 		}
@@ -4072,10 +3745,6 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 		if (!fs_info->scrub_wr_completion_workers)
 			goto fail_scrub_wr_completion_workers;
 
-		fs_info->scrub_nocow_workers =
-			btrfs_alloc_workqueue(fs_info, "scrubnc", flags, 1, 0);
-		if (!fs_info->scrub_nocow_workers)
-			goto fail_scrub_nocow_workers;
 		fs_info->scrub_parity_workers =
 			btrfs_alloc_workqueue(fs_info, "scrubparity", flags,
 					      max_active, 2);
@@ -4086,8 +3755,6 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 	return 0;
 
 fail_scrub_parity_workers:
-	btrfs_destroy_workqueue(fs_info->scrub_nocow_workers);
-fail_scrub_nocow_workers:
 	btrfs_destroy_workqueue(fs_info->scrub_wr_completion_workers);
 fail_scrub_wr_completion_workers:
 	btrfs_destroy_workqueue(fs_info->scrub_workers);
@@ -4100,7 +3767,6 @@ static noinline_for_stack void scrub_workers_put(struct btrfs_fs_info *fs_info)
 	if (--fs_info->scrub_workers_refcnt == 0) {
 		btrfs_destroy_workqueue(fs_info->scrub_workers);
 		btrfs_destroy_workqueue(fs_info->scrub_wr_completion_workers);
-		btrfs_destroy_workqueue(fs_info->scrub_nocow_workers);
 		btrfs_destroy_workqueue(fs_info->scrub_parity_workers);
 	}
 	WARN_ON(fs_info->scrub_workers_refcnt < 0);
@@ -4113,7 +3779,6 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	struct scrub_ctx *sctx;
 	int ret;
 	struct btrfs_device *dev;
-	struct rcu_string *name;
 
 	if (btrfs_fs_closing(fs_info))
 		return -EINVAL;
@@ -4167,11 +3832,8 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	if (!is_dev_replace && !readonly &&
 	    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		rcu_read_lock();
-		name = rcu_dereference(dev->name);
-		btrfs_err(fs_info, "scrub: device %s is not writable",
-			  name->str);
-		rcu_read_unlock();
+		btrfs_err_in_rcu(fs_info, "scrub: device %s is not writable",
+				rcu_str_deref(dev->name));
 		return -EROFS;
 	}
 
@@ -4359,330 +4021,3 @@ static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
 	*extent_dev = bbio->stripes[0].dev;
 	btrfs_put_bbio(bbio);
 }
-
-static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
-			    int mirror_num, u64 physical_for_dev_replace)
-{
-	struct scrub_copy_nocow_ctx *nocow_ctx;
-	struct btrfs_fs_info *fs_info = sctx->fs_info;
-
-	nocow_ctx = kzalloc(sizeof(*nocow_ctx), GFP_NOFS);
-	if (!nocow_ctx) {
-		spin_lock(&sctx->stat_lock);
-		sctx->stat.malloc_errors++;
-		spin_unlock(&sctx->stat_lock);
-		return -ENOMEM;
-	}
-
-	scrub_pending_trans_workers_inc(sctx);
-
-	nocow_ctx->sctx = sctx;
-	nocow_ctx->logical = logical;
-	nocow_ctx->len = len;
-	nocow_ctx->mirror_num = mirror_num;
-	nocow_ctx->physical_for_dev_replace = physical_for_dev_replace;
-	btrfs_init_work(&nocow_ctx->work, btrfs_scrubnc_helper,
-			copy_nocow_pages_worker, NULL, NULL);
-	INIT_LIST_HEAD(&nocow_ctx->inodes);
-	btrfs_queue_work(fs_info->scrub_nocow_workers,
-			 &nocow_ctx->work);
-
-	return 0;
-}
-
-static int record_inode_for_nocow(u64 inum, u64 offset, u64 root, void *ctx)
-{
-	struct scrub_copy_nocow_ctx *nocow_ctx = ctx;
-	struct scrub_nocow_inode *nocow_inode;
-
-	nocow_inode = kzalloc(sizeof(*nocow_inode), GFP_NOFS);
-	if (!nocow_inode)
-		return -ENOMEM;
-	nocow_inode->inum = inum;
-	nocow_inode->offset = offset;
-	nocow_inode->root = root;
-	list_add_tail(&nocow_inode->list, &nocow_ctx->inodes);
-	return 0;
-}
-
-#define COPY_COMPLETE 1
-
-static void copy_nocow_pages_worker(struct btrfs_work *work)
-{
-	struct scrub_copy_nocow_ctx *nocow_ctx =
-		container_of(work, struct scrub_copy_nocow_ctx, work);
-	struct scrub_ctx *sctx = nocow_ctx->sctx;
-	struct btrfs_fs_info *fs_info = sctx->fs_info;
-	struct btrfs_root *root = fs_info->extent_root;
-	u64 logical = nocow_ctx->logical;
-	u64 len = nocow_ctx->len;
-	int mirror_num = nocow_ctx->mirror_num;
-	u64 physical_for_dev_replace = nocow_ctx->physical_for_dev_replace;
-	int ret;
-	struct btrfs_trans_handle *trans = NULL;
-	struct btrfs_path *path;
-	int not_written = 0;
-
-	path = btrfs_alloc_path();
-	if (!path) {
-		spin_lock(&sctx->stat_lock);
-		sctx->stat.malloc_errors++;
-		spin_unlock(&sctx->stat_lock);
-		not_written = 1;
-		goto out;
-	}
-
-	trans = btrfs_join_transaction(root);
-	if (IS_ERR(trans)) {
-		not_written = 1;
-		goto out;
-	}
-
-	ret = iterate_inodes_from_logical(logical, fs_info, path,
-			record_inode_for_nocow, nocow_ctx, false);
-	if (ret != 0 && ret != -ENOENT) {
-		btrfs_warn(fs_info,
-			   "iterate_inodes_from_logical() failed: log %llu, phys %llu, len %llu, mir %u, ret %d",
-			   logical, physical_for_dev_replace, len, mirror_num,
-			   ret);
-		not_written = 1;
-		goto out;
-	}
-
-	btrfs_end_transaction(trans);
-	trans = NULL;
-	while (!list_empty(&nocow_ctx->inodes)) {
-		struct scrub_nocow_inode *entry;
-		entry = list_first_entry(&nocow_ctx->inodes,
-					 struct scrub_nocow_inode,
-					 list);
-		list_del_init(&entry->list);
-		ret = copy_nocow_pages_for_inode(entry->inum, entry->offset,
-						 entry->root, nocow_ctx);
-		kfree(entry);
-		if (ret == COPY_COMPLETE) {
-			ret = 0;
-			break;
-		} else if (ret) {
-			break;
-		}
-	}
-out:
-	while (!list_empty(&nocow_ctx->inodes)) {
-		struct scrub_nocow_inode *entry;
-		entry = list_first_entry(&nocow_ctx->inodes,
-					 struct scrub_nocow_inode,
-					 list);
-		list_del_init(&entry->list);
-		kfree(entry);
-	}
-	if (trans && !IS_ERR(trans))
-		btrfs_end_transaction(trans);
-	if (not_written)
-		btrfs_dev_replace_stats_inc(&fs_info->dev_replace.
-					    num_uncorrectable_read_errors);
-
-	btrfs_free_path(path);
-	kfree(nocow_ctx);
-
-	scrub_pending_trans_workers_dec(sctx);
-}
-
-static int check_extent_to_block(struct btrfs_inode *inode, u64 start, u64 len,
-				 u64 logical)
-{
-	struct extent_state *cached_state = NULL;
-	struct btrfs_ordered_extent *ordered;
-	struct extent_io_tree *io_tree;
-	struct extent_map *em;
-	u64 lockstart = start, lockend = start + len - 1;
-	int ret = 0;
-
-	io_tree = &inode->io_tree;
-
-	lock_extent_bits(io_tree, lockstart, lockend, &cached_state);
-	ordered = btrfs_lookup_ordered_range(inode, lockstart, len);
-	if (ordered) {
-		btrfs_put_ordered_extent(ordered);
-		ret = 1;
-		goto out_unlock;
-	}
-
-	em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
-	if (IS_ERR(em)) {
-		ret = PTR_ERR(em);
-		goto out_unlock;
-	}
-
-	/*
-	 * This extent does not actually cover the logical extent anymore,
-	 * move on to the next inode.
-	 */
-	if (em->block_start > logical ||
-	    em->block_start + em->block_len < logical + len ||
-	    test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
-		free_extent_map(em);
-		ret = 1;
-		goto out_unlock;
-	}
-	free_extent_map(em);
-
-out_unlock:
-	unlock_extent_cached(io_tree, lockstart, lockend, &cached_state);
-	return ret;
-}
-
-static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
-				      struct scrub_copy_nocow_ctx *nocow_ctx)
-{
-	struct btrfs_fs_info *fs_info = nocow_ctx->sctx->fs_info;
-	struct btrfs_key key;
-	struct inode *inode;
-	struct page *page;
-	struct btrfs_root *local_root;
-	struct extent_io_tree *io_tree;
-	u64 physical_for_dev_replace;
-	u64 nocow_ctx_logical;
-	u64 len = nocow_ctx->len;
-	unsigned long index;
-	int srcu_index;
-	int ret = 0;
-	int err = 0;
-
-	key.objectid = root;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	srcu_index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	local_root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(local_root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
-		return PTR_ERR(local_root);
-	}
-
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.objectid = inum;
-	key.offset = 0;
-	inode = btrfs_iget(fs_info->sb, &key, local_root, NULL);
-	srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
-	if (IS_ERR(inode))
-		return PTR_ERR(inode);
-
-	/* Avoid truncate/dio/punch hole.. */
-	inode_lock(inode);
-	inode_dio_wait(inode);
-
-	physical_for_dev_replace = nocow_ctx->physical_for_dev_replace;
-	io_tree = &BTRFS_I(inode)->io_tree;
-	nocow_ctx_logical = nocow_ctx->logical;
-
-	ret = check_extent_to_block(BTRFS_I(inode), offset, len,
-			nocow_ctx_logical);
-	if (ret) {
-		ret = ret > 0 ? 0 : ret;
-		goto out;
-	}
-
-	while (len >= PAGE_SIZE) {
-		index = offset >> PAGE_SHIFT;
-again:
-		page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
-		if (!page) {
-			btrfs_err(fs_info, "find_or_create_page() failed");
-			ret = -ENOMEM;
-			goto out;
-		}
-
-		if (PageUptodate(page)) {
-			if (PageDirty(page))
-				goto next_page;
-		} else {
-			ClearPageError(page);
-			err = extent_read_full_page(io_tree, page,
-							   btrfs_get_extent,
-							   nocow_ctx->mirror_num);
-			if (err) {
-				ret = err;
-				goto next_page;
-			}
-
-			lock_page(page);
-			/*
-			 * If the page has been remove from the page cache,
-			 * the data on it is meaningless, because it may be
-			 * old one, the new data may be written into the new
-			 * page in the page cache.
-			 */
-			if (page->mapping != inode->i_mapping) {
-				unlock_page(page);
-				put_page(page);
-				goto again;
-			}
-			if (!PageUptodate(page)) {
-				ret = -EIO;
-				goto next_page;
-			}
-		}
-
-		ret = check_extent_to_block(BTRFS_I(inode), offset, len,
-					    nocow_ctx_logical);
-		if (ret) {
-			ret = ret > 0 ? 0 : ret;
-			goto next_page;
-		}
-
-		err = write_page_nocow(nocow_ctx->sctx,
-				       physical_for_dev_replace, page);
-		if (err)
-			ret = err;
-next_page:
-		unlock_page(page);
-		put_page(page);
-
-		if (ret)
-			break;
-
-		offset += PAGE_SIZE;
-		physical_for_dev_replace += PAGE_SIZE;
-		nocow_ctx_logical += PAGE_SIZE;
-		len -= PAGE_SIZE;
-	}
-	ret = COPY_COMPLETE;
-out:
-	inode_unlock(inode);
-	iput(inode);
-	return ret;
-}
-
-static int write_page_nocow(struct scrub_ctx *sctx,
-			    u64 physical_for_dev_replace, struct page *page)
-{
-	struct bio *bio;
-	struct btrfs_device *dev;
-
-	dev = sctx->wr_tgtdev;
-	if (!dev)
-		return -EIO;
-	if (!dev->bdev) {
-		btrfs_warn_rl(dev->fs_info,
-			"scrub write_page_nocow(bdev == NULL) is unexpected");
-		return -EIO;
-	}
-	bio = btrfs_io_bio_alloc(1);
-	bio->bi_iter.bi_size = 0;
-	bio->bi_iter.bi_sector = physical_for_dev_replace >> 9;
-	bio_set_dev(bio, dev->bdev);
-	bio->bi_opf = REQ_OP_WRITE | REQ_SYNC;
-	/* bio_add_page won't fail on a freshly allocated bio */
-	bio_add_page(bio, page, PAGE_SIZE, 0);
-
-	if (btrfsic_submit_bio_wait(bio)) {
-		bio_put(bio);
-		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
-		return -EIO;
-	}
-
-	bio_put(bio);
-	return 0;
-}

fs/btrfs/send.c

@@ -100,6 +100,7 @@ struct send_ctx {
 	u64 cur_inode_rdev;
 	u64 cur_inode_last_extent;
 	u64 cur_inode_next_write_offset;
+	bool ignore_cur_inode;
 
 	u64 send_progress;
 
@@ -1500,7 +1501,7 @@ static int read_symlink(struct btrfs_root *root,
 	BUG_ON(compression);
 
 	off = btrfs_file_extent_inline_start(ei);
-	len = btrfs_file_extent_inline_len(path->nodes[0], path->slots[0], ei);
+	len = btrfs_file_extent_ram_bytes(path->nodes[0], ei);
 
 	ret = fs_path_add_from_extent_buffer(dest, path->nodes[0], off, len);
 
@@ -5006,6 +5007,15 @@ static int send_hole(struct send_ctx *sctx, u64 end)
 	u64 len;
 	int ret = 0;
 
+	/*
+	 * A hole that starts at EOF or beyond it. Since we do not yet support
+	 * fallocate (for extent preallocation and hole punching), sending a
+	 * write of zeroes starting at EOF or beyond would later require issuing
+	 * a truncate operation which would undo the write and achieve nothing.
+	 */
+	if (offset >= sctx->cur_inode_size)
+		return 0;
+
 	if (sctx->flags & BTRFS_SEND_FLAG_NO_FILE_DATA)
 		return send_update_extent(sctx, offset, end - offset);
 
@@ -5160,7 +5170,7 @@ static int clone_range(struct send_ctx *sctx,
 		ei = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 		type = btrfs_file_extent_type(leaf, ei);
 		if (type == BTRFS_FILE_EXTENT_INLINE) {
-			ext_len = btrfs_file_extent_inline_len(leaf, slot, ei);
+			ext_len = btrfs_file_extent_ram_bytes(leaf, ei);
 			ext_len = PAGE_ALIGN(ext_len);
 		} else {
 			ext_len = btrfs_file_extent_num_bytes(leaf, ei);
@@ -5236,8 +5246,7 @@ static int send_write_or_clone(struct send_ctx *sctx,
 			struct btrfs_file_extent_item);
 	type = btrfs_file_extent_type(path->nodes[0], ei);
 	if (type == BTRFS_FILE_EXTENT_INLINE) {
-		len = btrfs_file_extent_inline_len(path->nodes[0],
-						   path->slots[0], ei);
+		len = btrfs_file_extent_ram_bytes(path->nodes[0], ei);
 		/*
 		 * it is possible the inline item won't cover the whole page,
 		 * but there may be items after this page.  Make
@@ -5375,7 +5384,7 @@ static int is_extent_unchanged(struct send_ctx *sctx,
 		}
 
 		if (right_type == BTRFS_FILE_EXTENT_INLINE) {
-			right_len = btrfs_file_extent_inline_len(eb, slot, ei);
+			right_len = btrfs_file_extent_ram_bytes(eb, ei);
 			right_len = PAGE_ALIGN(right_len);
 		} else {
 			right_len = btrfs_file_extent_num_bytes(eb, ei);
@@ -5496,8 +5505,7 @@ static int get_last_extent(struct send_ctx *sctx, u64 offset)
 			    struct btrfs_file_extent_item);
 	type = btrfs_file_extent_type(path->nodes[0], fi);
 	if (type == BTRFS_FILE_EXTENT_INLINE) {
-		u64 size = btrfs_file_extent_inline_len(path->nodes[0],
-							path->slots[0], fi);
+		u64 size = btrfs_file_extent_ram_bytes(path->nodes[0], fi);
 		extent_end = ALIGN(key.offset + size,
 				   sctx->send_root->fs_info->sectorsize);
 	} else {
@@ -5560,7 +5568,7 @@ static int range_is_hole_in_parent(struct send_ctx *sctx,
 		fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 		if (btrfs_file_extent_type(leaf, fi) ==
 		    BTRFS_FILE_EXTENT_INLINE) {
-			u64 size = btrfs_file_extent_inline_len(leaf, slot, fi);
+			u64 size = btrfs_file_extent_ram_bytes(leaf, fi);
 
 			extent_end = ALIGN(key.offset + size,
 					   root->fs_info->sectorsize);
@@ -5606,8 +5614,7 @@ static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path,
 			    struct btrfs_file_extent_item);
 	type = btrfs_file_extent_type(path->nodes[0], fi);
 	if (type == BTRFS_FILE_EXTENT_INLINE) {
-		u64 size = btrfs_file_extent_inline_len(path->nodes[0],
-							path->slots[0], fi);
+		u64 size = btrfs_file_extent_ram_bytes(path->nodes[0], fi);
 		extent_end = ALIGN(key->offset + size,
 				   sctx->send_root->fs_info->sectorsize);
 	} else {
@@ -5799,6 +5806,9 @@ static int finish_inode_if_needed(struct send_ctx *sctx, int at_end)
 	int pending_move = 0;
 	int refs_processed = 0;
 
+	if (sctx->ignore_cur_inode)
+		return 0;
+
 	ret = process_recorded_refs_if_needed(sctx, at_end, &pending_move,
 					      &refs_processed);
 	if (ret < 0)
@@ -5917,6 +5927,93 @@ out:
 	return ret;
 }
 
+struct parent_paths_ctx {
+	struct list_head *refs;
+	struct send_ctx *sctx;
+};
+
+static int record_parent_ref(int num, u64 dir, int index, struct fs_path *name,
+			     void *ctx)
+{
+	struct parent_paths_ctx *ppctx = ctx;
+
+	return record_ref(ppctx->sctx->parent_root, dir, name, ppctx->sctx,
+			  ppctx->refs);
+}
+
+/*
+ * Issue unlink operations for all paths of the current inode found in the
+ * parent snapshot.
+ */
+static int btrfs_unlink_all_paths(struct send_ctx *sctx)
+{
+	LIST_HEAD(deleted_refs);
+	struct btrfs_path *path;
+	struct btrfs_key key;
+	struct parent_paths_ctx ctx;
+	int ret;
+
+	path = alloc_path_for_send();
+	if (!path)
+		return -ENOMEM;
+
+	key.objectid = sctx->cur_ino;
+	key.type = BTRFS_INODE_REF_KEY;
+	key.offset = 0;
+	ret = btrfs_search_slot(NULL, sctx->parent_root, &key, path, 0, 0);
+	if (ret < 0)
+		goto out;
+
+	ctx.refs = &deleted_refs;
+	ctx.sctx = sctx;
+
+	while (true) {
+		struct extent_buffer *eb = path->nodes[0];
+		int slot = path->slots[0];
+
+		if (slot >= btrfs_header_nritems(eb)) {
+			ret = btrfs_next_leaf(sctx->parent_root, path);
+			if (ret < 0)
+				goto out;
+			else if (ret > 0)
+				break;
+			continue;
+		}
+
+		btrfs_item_key_to_cpu(eb, &key, slot);
+		if (key.objectid != sctx->cur_ino)
+			break;
+		if (key.type != BTRFS_INODE_REF_KEY &&
+		    key.type != BTRFS_INODE_EXTREF_KEY)
+			break;
+
+		ret = iterate_inode_ref(sctx->parent_root, path, &key, 1,
+					record_parent_ref, &ctx);
+		if (ret < 0)
+			goto out;
+
+		path->slots[0]++;
+	}
+
+	while (!list_empty(&deleted_refs)) {
+		struct recorded_ref *ref;
+
+		ref = list_first_entry(&deleted_refs, struct recorded_ref, list);
+		ret = send_unlink(sctx, ref->full_path);
+		if (ret < 0)
+			goto out;
+		fs_path_free(ref->full_path);
+		list_del(&ref->list);
+		kfree(ref);
+	}
+	ret = 0;
+out:
+	btrfs_free_path(path);
+	if (ret)
+		__free_recorded_refs(&deleted_refs);
+	return ret;
+}
+
 static int changed_inode(struct send_ctx *sctx,
 			 enum btrfs_compare_tree_result result)
 {
@@ -5931,6 +6028,7 @@ static int changed_inode(struct send_ctx *sctx,
 	sctx->cur_inode_new_gen = 0;
 	sctx->cur_inode_last_extent = (u64)-1;
 	sctx->cur_inode_next_write_offset = 0;
+	sctx->ignore_cur_inode = false;
 
 	/*
 	 * Set send_progress to current inode. This will tell all get_cur_xxx
@@ -5971,6 +6069,33 @@ static int changed_inode(struct send_ctx *sctx,
 			sctx->cur_inode_new_gen = 1;
 	}
 
+	/*
+	 * Normally we do not find inodes with a link count of zero (orphans)
+	 * because the most common case is to create a snapshot and use it
+	 * for a send operation. However other less common use cases involve
+	 * using a subvolume and send it after turning it to RO mode just
+	 * after deleting all hard links of a file while holding an open
+	 * file descriptor against it or turning a RO snapshot into RW mode,
+	 * keep an open file descriptor against a file, delete it and then
+	 * turn the snapshot back to RO mode before using it for a send
+	 * operation. So if we find such cases, ignore the inode and all its
+	 * items completely if it's a new inode, or if it's a changed inode
+	 * make sure all its previous paths (from the parent snapshot) are all
+	 * unlinked and all other the inode items are ignored.
+	 */
+	if (result == BTRFS_COMPARE_TREE_NEW ||
+	    result == BTRFS_COMPARE_TREE_CHANGED) {
+		u32 nlinks;
+
+		nlinks = btrfs_inode_nlink(sctx->left_path->nodes[0], left_ii);
+		if (nlinks == 0) {
+			sctx->ignore_cur_inode = true;
+			if (result == BTRFS_COMPARE_TREE_CHANGED)
+				ret = btrfs_unlink_all_paths(sctx);
+			goto out;
+		}
+	}
+
 	if (result == BTRFS_COMPARE_TREE_NEW) {
 		sctx->cur_inode_gen = left_gen;
 		sctx->cur_inode_new = 1;
@@ -6309,15 +6434,17 @@ static int changed_cb(struct btrfs_path *left_path,
 	    key->objectid == BTRFS_FREE_SPACE_OBJECTID)
 		goto out;
 
-	if (key->type == BTRFS_INODE_ITEM_KEY)
+	if (key->type == BTRFS_INODE_ITEM_KEY) {
 		ret = changed_inode(sctx, result);
-	else if (key->type == BTRFS_INODE_REF_KEY ||
-		 key->type == BTRFS_INODE_EXTREF_KEY)
-		ret = changed_ref(sctx, result);
-	else if (key->type == BTRFS_XATTR_ITEM_KEY)
-		ret = changed_xattr(sctx, result);
-	else if (key->type == BTRFS_EXTENT_DATA_KEY)
-		ret = changed_extent(sctx, result);
+	} else if (!sctx->ignore_cur_inode) {
+		if (key->type == BTRFS_INODE_REF_KEY ||
+		    key->type == BTRFS_INODE_EXTREF_KEY)
+			ret = changed_ref(sctx, result);
+		else if (key->type == BTRFS_XATTR_ITEM_KEY)
+			ret = changed_xattr(sctx, result);
+		else if (key->type == BTRFS_EXTENT_DATA_KEY)
+			ret = changed_extent(sctx, result);
+	}
 
 out:
 	return ret;
@@ -6328,7 +6455,6 @@ static int full_send_tree(struct send_ctx *sctx)
 	int ret;
 	struct btrfs_root *send_root = sctx->send_root;
 	struct btrfs_key key;
-	struct btrfs_key found_key;
 	struct btrfs_path *path;
 	struct extent_buffer *eb;
 	int slot;
@@ -6350,17 +6476,13 @@ static int full_send_tree(struct send_ctx *sctx)
 	while (1) {
 		eb = path->nodes[0];
 		slot = path->slots[0];
-		btrfs_item_key_to_cpu(eb, &found_key, slot);
+		btrfs_item_key_to_cpu(eb, &key, slot);
 
-		ret = changed_cb(path, NULL, &found_key,
+		ret = changed_cb(path, NULL, &key,
 				 BTRFS_COMPARE_TREE_NEW, sctx);
 		if (ret < 0)
 			goto out;
 
-		key.objectid = found_key.objectid;
-		key.type = found_key.type;
-		key.offset = found_key.offset + 1;
-
 		ret = btrfs_next_item(send_root, path);
 		if (ret < 0)
 			goto out;

fs/btrfs/struct-funcs.c

@@ -3,7 +3,6 @@
  * Copyright (C) 2007 Oracle.  All rights reserved.
  */
 
-#include <linux/highmem.h>
 #include <asm/unaligned.h>
 
 #include "ctree.h"

fs/btrfs/super.c

@@ -5,7 +5,6 @@
 
 #include <linux/blkdev.h>
 #include <linux/module.h>
-#include <linux/buffer_head.h>
 #include <linux/fs.h>
 #include <linux/pagemap.h>
 #include <linux/highmem.h>
@@ -15,8 +14,6 @@
 #include <linux/string.h>
 #include <linux/backing-dev.h>
 #include <linux/mount.h>
-#include <linux/mpage.h>
-#include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/statfs.h>
 #include <linux/compat.h>
@@ -468,9 +465,8 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 		case Opt_subvolrootid:
 		case Opt_device:
 			/*
-			 * These are parsed by btrfs_parse_subvol_options
-			 * and btrfs_parse_early_options
-			 * and can be happily ignored here.
+			 * These are parsed by btrfs_parse_subvol_options or
+			 * btrfs_parse_device_options and can be ignored here.
 			 */
 			break;
 		case Opt_nodatasum:
@@ -760,6 +756,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 		case Opt_recovery:
 			btrfs_warn(info,
 				   "'recovery' is deprecated, use 'usebackuproot' instead");
+			/* fall through */
 		case Opt_usebackuproot:
 			btrfs_info(info,
 				   "trying to use backup root at mount time");
@@ -885,13 +882,16 @@ out:
  * All other options will be parsed on much later in the mount process and
  * only when we need to allocate a new super block.
  */
-static int btrfs_parse_early_options(const char *options, fmode_t flags,
-		void *holder, struct btrfs_fs_devices **fs_devices)
+static int btrfs_parse_device_options(const char *options, fmode_t flags,
+				      void *holder)
 {
 	substring_t args[MAX_OPT_ARGS];
 	char *device_name, *opts, *orig, *p;
+	struct btrfs_device *device = NULL;
 	int error = 0;
 
+	lockdep_assert_held(&uuid_mutex);
+
 	if (!options)
 		return 0;
 
@@ -917,11 +917,13 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 				error = -ENOMEM;
 				goto out;
 			}
-			error = btrfs_scan_one_device(device_name,
-					flags, holder, fs_devices);
+			device = btrfs_scan_one_device(device_name, flags,
+					holder);
 			kfree(device_name);
-			if (error)
+			if (IS_ERR(device)) {
+				error = PTR_ERR(device);
 				goto out;
+			}
 		}
 	}
 
@@ -935,8 +937,8 @@ out:
  *
  * The value is later passed to mount_subvol()
  */
-static int btrfs_parse_subvol_options(const char *options, fmode_t flags,
-		char **subvol_name, u64 *subvol_objectid)
+static int btrfs_parse_subvol_options(const char *options, char **subvol_name,
+		u64 *subvol_objectid)
 {
 	substring_t args[MAX_OPT_ARGS];
 	char *opts, *orig, *p;
@@ -948,7 +950,7 @@ static int btrfs_parse_subvol_options(const char *options, fmode_t flags,
 
 	/*
 	 * strsep changes the string, duplicate it because
-	 * btrfs_parse_early_options gets called later
+	 * btrfs_parse_device_options gets called later
 	 */
 	opts = kstrdup(options, GFP_KERNEL);
 	if (!opts)
@@ -1517,6 +1519,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 {
 	struct block_device *bdev = NULL;
 	struct super_block *s;
+	struct btrfs_device *device = NULL;
 	struct btrfs_fs_devices *fs_devices = NULL;
 	struct btrfs_fs_info *fs_info = NULL;
 	struct security_mnt_opts new_sec_opts;
@@ -1526,12 +1529,6 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 	if (!(flags & SB_RDONLY))
 		mode |= FMODE_WRITE;
 
-	error = btrfs_parse_early_options(data, mode, fs_type,
-					  &fs_devices);
-	if (error) {
-		return ERR_PTR(error);
-	}
-
 	security_init_mnt_opts(&new_sec_opts);
 	if (data) {
 		error = parse_security_options(data, &new_sec_opts);
@@ -1539,10 +1536,6 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 			return ERR_PTR(error);
 	}
 
-	error = btrfs_scan_one_device(device_name, mode, fs_type, &fs_devices);
-	if (error)
-		goto error_sec_opts;
-
 	/*
 	 * Setup a dummy root and fs_info for test/set super.  This is because
 	 * we don't actually fill this stuff out until open_ctree, but we need
@@ -1555,8 +1548,6 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 		goto error_sec_opts;
 	}
 
-	fs_info->fs_devices = fs_devices;
-
 	fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL);
 	fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL);
 	security_init_mnt_opts(&fs_info->security_opts);
@@ -1565,7 +1556,25 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 		goto error_fs_info;
 	}
 
+	mutex_lock(&uuid_mutex);
+	error = btrfs_parse_device_options(data, mode, fs_type);
+	if (error) {
+		mutex_unlock(&uuid_mutex);
+		goto error_fs_info;
+	}
+
+	device = btrfs_scan_one_device(device_name, mode, fs_type);
+	if (IS_ERR(device)) {
+		mutex_unlock(&uuid_mutex);
+		error = PTR_ERR(device);
+		goto error_fs_info;
+	}
+
+	fs_devices = device->fs_devices;
+	fs_info->fs_devices = fs_devices;
+
 	error = btrfs_open_devices(fs_devices, mode, fs_type);
+	mutex_unlock(&uuid_mutex);
 	if (error)
 		goto error_fs_info;
 
@@ -1650,8 +1659,8 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 	if (!(flags & SB_RDONLY))
 		mode |= FMODE_WRITE;
 
-	error = btrfs_parse_subvol_options(data, mode,
-					  &subvol_name, &subvol_objectid);
+	error = btrfs_parse_subvol_options(data, &subvol_name,
+					&subvol_objectid);
 	if (error) {
 		kfree(subvol_name);
 		return ERR_PTR(error);
@@ -2098,14 +2107,9 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 				btrfs_account_ro_block_groups_free_space(found);
 
 			for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
-				if (!list_empty(&found->block_groups[i])) {
-					switch (i) {
-					case BTRFS_RAID_DUP:
-					case BTRFS_RAID_RAID1:
-					case BTRFS_RAID_RAID10:
-						factor = 2;
-					}
-				}
+				if (!list_empty(&found->block_groups[i]))
+					factor = btrfs_bg_type_to_factor(
+						btrfs_raid_array[i].bg_flag);
 			}
 		}
 
@@ -2222,7 +2226,7 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 				unsigned long arg)
 {
 	struct btrfs_ioctl_vol_args *vol;
-	struct btrfs_fs_devices *fs_devices;
+	struct btrfs_device *device = NULL;
 	int ret = -ENOTTY;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -2234,15 +2238,24 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 
 	switch (cmd) {
 	case BTRFS_IOC_SCAN_DEV:
-		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
-					    &btrfs_root_fs_type, &fs_devices);
+		mutex_lock(&uuid_mutex);
+		device = btrfs_scan_one_device(vol->name, FMODE_READ,
+					       &btrfs_root_fs_type);
+		ret = PTR_ERR_OR_ZERO(device);
+		mutex_unlock(&uuid_mutex);
 		break;
 	case BTRFS_IOC_DEVICES_READY:
-		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
-					    &btrfs_root_fs_type, &fs_devices);
-		if (ret)
+		mutex_lock(&uuid_mutex);
+		device = btrfs_scan_one_device(vol->name, FMODE_READ,
+					       &btrfs_root_fs_type);
+		if (IS_ERR(device)) {
+			mutex_unlock(&uuid_mutex);
+			ret = PTR_ERR(device);
 			break;
-		ret = !(fs_devices->num_devices == fs_devices->total_devices);
+		}
+		ret = !(device->fs_devices->num_devices ==
+			device->fs_devices->total_devices);
+		mutex_unlock(&uuid_mutex);
 		break;
 	case BTRFS_IOC_GET_SUPPORTED_FEATURES:
 		ret = btrfs_ioctl_get_supported_features((void __user*)arg);
@@ -2290,7 +2303,6 @@ static int btrfs_show_devname(struct seq_file *m, struct dentry *root)
 	struct btrfs_fs_devices *cur_devices;
 	struct btrfs_device *dev, *first_dev = NULL;
 	struct list_head *head;
-	struct rcu_string *name;
 
 	/*
 	 * Lightweight locking of the devices. We should not need
@@ -2314,12 +2326,10 @@ static int btrfs_show_devname(struct seq_file *m, struct dentry *root)
 		cur_devices = cur_devices->seed;
 	}
 
-	if (first_dev) {
-		name = rcu_dereference(first_dev->name);
-		seq_escape(m, name->str, " \t\n\\");
-	} else {
+	if (first_dev)
+		seq_escape(m, rcu_str_deref(first_dev->name), " \t\n\\");
+	else
 		WARN_ON(1);
-	}
 	rcu_read_unlock();
 	return 0;
 }
@@ -2331,7 +2341,6 @@ static const struct super_operations btrfs_super_ops = {
 	.sync_fs	= btrfs_sync_fs,
 	.show_options	= btrfs_show_options,
 	.show_devname	= btrfs_show_devname,
-	.write_inode	= btrfs_write_inode,
 	.alloc_inode	= btrfs_alloc_inode,
 	.destroy_inode	= btrfs_destroy_inode,
 	.statfs		= btrfs_statfs,
@@ -2369,7 +2378,7 @@ static __cold void btrfs_interface_exit(void)
 
 static void __init btrfs_print_mod_info(void)
 {
-	pr_info("Btrfs loaded, crc32c=%s"
+	static const char options[] = ""
 #ifdef CONFIG_BTRFS_DEBUG
 			", debug=on"
 #endif
@@ -2382,8 +2391,8 @@ static void __init btrfs_print_mod_info(void)
 #ifdef CONFIG_BTRFS_FS_REF_VERIFY
 			", ref-verify=on"
 #endif
-			"\n",
-			crc32c_impl());
+			;
+	pr_info("Btrfs loaded, crc32c=%s%s\n", crc32c_impl(), options);
 }
 
 static int __init init_btrfs_fs(void)

fs/btrfs/sysfs.c

@@ -7,10 +7,8 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/completion.h>
-#include <linux/buffer_head.h>
 #include <linux/kobject.h>
 #include <linux/bug.h>
-#include <linux/genhd.h>
 #include <linux/debugfs.h>
 
 #include "ctree.h"

fs/btrfs/tests/qgroup-tests.c

@@ -216,7 +216,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 	btrfs_init_dummy_trans(&trans, fs_info);
 
 	test_msg("qgroup basic add");
-	ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FS_TREE_OBJECTID);
+	ret = btrfs_create_qgroup(&trans, BTRFS_FS_TREE_OBJECTID);
 	if (ret) {
 		test_err("couldn't create a qgroup %d", ret);
 		return ret;
@@ -249,8 +249,8 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
-					  nodesize, old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, nodesize, nodesize, old_roots,
+					  new_roots);
 	if (ret) {
 		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
@@ -285,8 +285,8 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
-					  nodesize, old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, nodesize, nodesize, old_roots,
+					  new_roots);
 	if (ret) {
 		test_err("couldn't account space for a qgroup %d", ret);
 		return -EINVAL;
@@ -322,7 +322,7 @@ static int test_multiple_refs(struct btrfs_root *root,
 	 * We have BTRFS_FS_TREE_OBJECTID created already from the
 	 * previous test.
 	 */
-	ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FIRST_FREE_OBJECTID);
+	ret = btrfs_create_qgroup(&trans, BTRFS_FIRST_FREE_OBJECTID);
 	if (ret) {
 		test_err("couldn't create a qgroup %d", ret);
 		return ret;
@@ -350,8 +350,8 @@ static int test_multiple_refs(struct btrfs_root *root,
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
-					  nodesize, old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, nodesize, nodesize, old_roots,
+					  new_roots);
 	if (ret) {
 		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
@@ -385,8 +385,8 @@ static int test_multiple_refs(struct btrfs_root *root,
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
-					  nodesize, old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, nodesize, nodesize, old_roots,
+					  new_roots);
 	if (ret) {
 		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
@@ -426,8 +426,8 @@ static int test_multiple_refs(struct btrfs_root *root,
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
-					  nodesize, old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, nodesize, nodesize, old_roots,
+					  new_roots);
 	if (ret) {
 		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;

fs/btrfs/transaction.c

@@ -241,7 +241,7 @@ loop:
 	refcount_set(&cur_trans->use_count, 2);
 	atomic_set(&cur_trans->pending_ordered, 0);
 	cur_trans->flags = 0;
-	cur_trans->start_time = get_seconds();
+	cur_trans->start_time = ktime_get_seconds();
 
 	memset(&cur_trans->delayed_refs, 0, sizeof(cur_trans->delayed_refs));
 
@@ -680,7 +680,7 @@ btrfs_attach_transaction_barrier(struct btrfs_root *root)
 
 	trans = start_transaction(root, 0, TRANS_ATTACH,
 				  BTRFS_RESERVE_NO_FLUSH, true);
-	if (IS_ERR(trans) && PTR_ERR(trans) == -ENOENT)
+	if (trans == ERR_PTR(-ENOENT))
 		btrfs_wait_for_commit(root->fs_info, 0);
 
 	return trans;
@@ -1152,7 +1152,7 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans)
 	ret = btrfs_run_dev_replace(trans, fs_info);
 	if (ret)
 		return ret;
-	ret = btrfs_run_qgroups(trans, fs_info);
+	ret = btrfs_run_qgroups(trans);
 	if (ret)
 		return ret;
 
@@ -1355,8 +1355,7 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
 		goto out;
 
 	/* Now qgroup are all updated, we can inherit it to new qgroups */
-	ret = btrfs_qgroup_inherit(trans, fs_info,
-				   src->root_key.objectid, dst_objectid,
+	ret = btrfs_qgroup_inherit(trans, src->root_key.objectid, dst_objectid,
 				   inherit);
 	if (ret < 0)
 		goto out;
@@ -1574,7 +1573,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	/*
 	 * insert root back/forward references
 	 */
-	ret = btrfs_add_root_ref(trans, fs_info, objectid,
+	ret = btrfs_add_root_ref(trans, objectid,
 				 parent_root->root_key.objectid,
 				 btrfs_ino(BTRFS_I(parent_inode)), index,
 				 dentry->d_name.name, dentry->d_name.len);

fs/btrfs/transaction.h

@@ -48,7 +48,7 @@ struct btrfs_transaction {
 	int aborted;
 	struct list_head list;
 	struct extent_io_tree dirty_pages;
-	unsigned long start_time;
+	time64_t start_time;
 	wait_queue_head_t writer_wait;
 	wait_queue_head_t commit_wait;
 	wait_queue_head_t pending_wait;

fs/btrfs/tree-checker.c

@@ -19,6 +19,7 @@
 #include "tree-checker.h"
 #include "disk-io.h"
 #include "compression.h"
+#include "volumes.h"
 
 /*
  * Error message should follow the following format:
@@ -353,6 +354,102 @@ static int check_dir_item(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
+__printf(4, 5)
+__cold
+static void block_group_err(const struct btrfs_fs_info *fs_info,
+			    const struct extent_buffer *eb, int slot,
+			    const char *fmt, ...)
+{
+	struct btrfs_key key;
+	struct va_format vaf;
+	va_list args;
+
+	btrfs_item_key_to_cpu(eb, &key, slot);
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	btrfs_crit(fs_info,
+	"corrupt %s: root=%llu block=%llu slot=%d bg_start=%llu bg_len=%llu, %pV",
+		btrfs_header_level(eb) == 0 ? "leaf" : "node",
+		btrfs_header_owner(eb), btrfs_header_bytenr(eb), slot,
+		key.objectid, key.offset, &vaf);
+	va_end(args);
+}
+
+static int check_block_group_item(struct btrfs_fs_info *fs_info,
+				  struct extent_buffer *leaf,
+				  struct btrfs_key *key, int slot)
+{
+	struct btrfs_block_group_item bgi;
+	u32 item_size = btrfs_item_size_nr(leaf, slot);
+	u64 flags;
+	u64 type;
+
+	/*
+	 * Here we don't really care about alignment since extent allocator can
+	 * handle it.  We care more about the size, as if one block group is
+	 * larger than maximum size, it's must be some obvious corruption.
+	 */
+	if (key->offset > BTRFS_MAX_DATA_CHUNK_SIZE || key->offset == 0) {
+		block_group_err(fs_info, leaf, slot,
+			"invalid block group size, have %llu expect (0, %llu]",
+				key->offset, BTRFS_MAX_DATA_CHUNK_SIZE);
+		return -EUCLEAN;
+	}
+
+	if (item_size != sizeof(bgi)) {
+		block_group_err(fs_info, leaf, slot,
+			"invalid item size, have %u expect %zu",
+				item_size, sizeof(bgi));
+		return -EUCLEAN;
+	}
+
+	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
+			   sizeof(bgi));
+	if (btrfs_block_group_chunk_objectid(&bgi) !=
+	    BTRFS_FIRST_CHUNK_TREE_OBJECTID) {
+		block_group_err(fs_info, leaf, slot,
+		"invalid block group chunk objectid, have %llu expect %llu",
+				btrfs_block_group_chunk_objectid(&bgi),
+				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
+		return -EUCLEAN;
+	}
+
+	if (btrfs_block_group_used(&bgi) > key->offset) {
+		block_group_err(fs_info, leaf, slot,
+			"invalid block group used, have %llu expect [0, %llu)",
+				btrfs_block_group_used(&bgi), key->offset);
+		return -EUCLEAN;
+	}
+
+	flags = btrfs_block_group_flags(&bgi);
+	if (hweight64(flags & BTRFS_BLOCK_GROUP_PROFILE_MASK) > 1) {
+		block_group_err(fs_info, leaf, slot,
+"invalid profile flags, have 0x%llx (%lu bits set) expect no more than 1 bit set",
+			flags & BTRFS_BLOCK_GROUP_PROFILE_MASK,
+			hweight64(flags & BTRFS_BLOCK_GROUP_PROFILE_MASK));
+		return -EUCLEAN;
+	}
+
+	type = flags & BTRFS_BLOCK_GROUP_TYPE_MASK;
+	if (type != BTRFS_BLOCK_GROUP_DATA &&
+	    type != BTRFS_BLOCK_GROUP_METADATA &&
+	    type != BTRFS_BLOCK_GROUP_SYSTEM &&
+	    type != (BTRFS_BLOCK_GROUP_METADATA |
+			   BTRFS_BLOCK_GROUP_DATA)) {
+		block_group_err(fs_info, leaf, slot,
+"invalid type, have 0x%llx (%lu bits set) expect either 0x%llx, 0x%llx, 0x%llu or 0x%llx",
+			type, hweight64(type),
+			BTRFS_BLOCK_GROUP_DATA, BTRFS_BLOCK_GROUP_METADATA,
+			BTRFS_BLOCK_GROUP_SYSTEM,
+			BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA);
+		return -EUCLEAN;
+	}
+	return 0;
+}
+
 /*
  * Common point to switch the item-specific validation.
  */
@@ -374,6 +471,9 @@ static int check_leaf_item(struct btrfs_fs_info *fs_info,
 	case BTRFS_XATTR_ITEM_KEY:
 		ret = check_dir_item(fs_info, leaf, key, slot);
 		break;
+	case BTRFS_BLOCK_GROUP_ITEM_KEY:
+		ret = check_block_group_item(fs_info, leaf, key, slot);
+		break;
 	}
 	return ret;
 }
@@ -396,9 +496,22 @@ static int check_leaf(struct btrfs_fs_info *fs_info, struct extent_buffer *leaf,
 	 * skip this check for relocation trees.
 	 */
 	if (nritems == 0 && !btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_RELOC)) {
+		u64 owner = btrfs_header_owner(leaf);
 		struct btrfs_root *check_root;
 
-		key.objectid = btrfs_header_owner(leaf);
+		/* These trees must never be empty */
+		if (owner == BTRFS_ROOT_TREE_OBJECTID ||
+		    owner == BTRFS_CHUNK_TREE_OBJECTID ||
+		    owner == BTRFS_EXTENT_TREE_OBJECTID ||
+		    owner == BTRFS_DEV_TREE_OBJECTID ||
+		    owner == BTRFS_FS_TREE_OBJECTID ||
+		    owner == BTRFS_DATA_RELOC_TREE_OBJECTID) {
+			generic_err(fs_info, leaf, 0,
+			"invalid root, root %llu must never be empty",
+				    owner);
+			return -EUCLEAN;
+		}
+		key.objectid = owner;
 		key.type = BTRFS_ROOT_ITEM_KEY;
 		key.offset = (u64)-1;
 

fs/btrfs/tree-log.c

@@ -545,12 +545,8 @@ static noinline struct inode *read_one_inode(struct btrfs_root *root,
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 	inode = btrfs_iget(root->fs_info->sb, &key, root, NULL);
-	if (IS_ERR(inode)) {
+	if (IS_ERR(inode))
 		inode = NULL;
-	} else if (is_bad_inode(inode)) {
-		iput(inode);
-		inode = NULL;
-	}
 	return inode;
 }
 
@@ -597,7 +593,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 		if (btrfs_file_extent_disk_bytenr(eb, item) == 0)
 			nbytes = 0;
 	} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
-		size = btrfs_file_extent_inline_len(eb, slot, item);
+		size = btrfs_file_extent_ram_bytes(eb, item);
 		nbytes = btrfs_file_extent_ram_bytes(eb, item);
 		extent_end = ALIGN(start + size,
 				   fs_info->sectorsize);
@@ -685,7 +681,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 		 * as the owner of the file extent changed from log tree
 		 * (doesn't affect qgroup) to fs/file tree(affects qgroup)
 		 */
-		ret = btrfs_qgroup_trace_extent(trans, fs_info,
+		ret = btrfs_qgroup_trace_extent(trans,
 				btrfs_file_extent_disk_bytenr(eb, item),
 				btrfs_file_extent_disk_num_bytes(eb, item),
 				GFP_NOFS);
@@ -715,7 +711,6 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 				 * allocation tree
 				 */
 				ret = btrfs_alloc_logged_file_extent(trans,
-						fs_info,
 						root->root_key.objectid,
 						key->objectid, offset, &ins);
 				if (ret)
@@ -1291,6 +1286,46 @@ again:
 	return ret;
 }
 
+static int btrfs_inode_ref_exists(struct inode *inode, struct inode *dir,
+				  const u8 ref_type, const char *name,
+				  const int namelen)
+{
+	struct btrfs_key key;
+	struct btrfs_path *path;
+	const u64 parent_id = btrfs_ino(BTRFS_I(dir));
+	int ret;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	key.objectid = btrfs_ino(BTRFS_I(inode));
+	key.type = ref_type;
+	if (key.type == BTRFS_INODE_REF_KEY)
+		key.offset = parent_id;
+	else
+		key.offset = btrfs_extref_hash(parent_id, name, namelen);
+
+	ret = btrfs_search_slot(NULL, BTRFS_I(inode)->root, &key, path, 0, 0);
+	if (ret < 0)
+		goto out;
+	if (ret > 0) {
+		ret = 0;
+		goto out;
+	}
+	if (key.type == BTRFS_INODE_EXTREF_KEY)
+		ret = btrfs_find_name_in_ext_backref(path->nodes[0],
+						     path->slots[0], parent_id,
+						     name, namelen, NULL);
+	else
+		ret = btrfs_find_name_in_backref(path->nodes[0], path->slots[0],
+						 name, namelen, NULL);
+
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
 /*
  * replay one inode back reference item found in the log tree.
  * eb, slot and key refer to the buffer and key found in the log tree.
@@ -1400,6 +1435,32 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 				}
 			}
 
+			/*
+			 * If a reference item already exists for this inode
+			 * with the same parent and name, but different index,
+			 * drop it and the corresponding directory index entries
+			 * from the parent before adding the new reference item
+			 * and dir index entries, otherwise we would fail with
+			 * -EEXIST returned from btrfs_add_link() below.
+			 */
+			ret = btrfs_inode_ref_exists(inode, dir, key->type,
+						     name, namelen);
+			if (ret > 0) {
+				ret = btrfs_unlink_inode(trans, root,
+							 BTRFS_I(dir),
+							 BTRFS_I(inode),
+							 name, namelen);
+				/*
+				 * If we dropped the link count to 0, bump it so
+				 * that later the iput() on the inode will not
+				 * free it. We will fixup the link count later.
+				 */
+				if (!ret && inode->i_nlink == 0)
+					inc_nlink(inode);
+			}
+			if (ret < 0)
+				goto out;
+
 			/* insert our name */
 			ret = btrfs_add_link(trans, BTRFS_I(dir),
 					BTRFS_I(inode),
@@ -2120,7 +2181,7 @@ again:
 						     dir_key->offset,
 						     name, name_len, 0);
 		}
-		if (!log_di || (IS_ERR(log_di) && PTR_ERR(log_di) == -ENOENT)) {
+		if (!log_di || log_di == ERR_PTR(-ENOENT)) {
 			btrfs_dir_item_key_to_cpu(eb, di, &location);
 			btrfs_release_path(path);
 			btrfs_release_path(log_path);
@@ -2933,7 +2994,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	/* bail out if we need to do a full commit */
 	if (btrfs_need_log_full_commit(fs_info, trans)) {
 		ret = -EAGAIN;
-		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&root->log_mutex);
 		goto out;
 	}
@@ -2951,7 +3011,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	if (ret) {
 		blk_finish_plug(&plug);
 		btrfs_abort_transaction(trans, ret);
-		btrfs_free_logged_extents(log, log_transid);
 		btrfs_set_log_full_commit(fs_info, trans);
 		mutex_unlock(&root->log_mutex);
 		goto out;
@@ -3002,7 +3061,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 			goto out;
 		}
 		btrfs_wait_tree_log_extents(log, mark);
-		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
 		ret = -EAGAIN;
 		goto out;
@@ -3020,7 +3078,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	if (atomic_read(&log_root_tree->log_commit[index2])) {
 		blk_finish_plug(&plug);
 		ret = btrfs_wait_tree_log_extents(log, mark);
-		btrfs_wait_logged_extents(trans, log, log_transid);
 		wait_log_commit(log_root_tree,
 				root_log_ctx.log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
@@ -3045,7 +3102,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	if (btrfs_need_log_full_commit(fs_info, trans)) {
 		blk_finish_plug(&plug);
 		btrfs_wait_tree_log_extents(log, mark);
-		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
 		ret = -EAGAIN;
 		goto out_wake_log_root;
@@ -3058,7 +3114,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	if (ret) {
 		btrfs_set_log_full_commit(fs_info, trans);
 		btrfs_abort_transaction(trans, ret);
-		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
 		goto out_wake_log_root;
 	}
@@ -3068,11 +3123,9 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 						  EXTENT_NEW | EXTENT_DIRTY);
 	if (ret) {
 		btrfs_set_log_full_commit(fs_info, trans);
-		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
 		goto out_wake_log_root;
 	}
-	btrfs_wait_logged_extents(trans, log, log_transid);
 
 	btrfs_set_super_log_root(fs_info->super_for_commit,
 				 log_root_tree->node->start);
@@ -3159,14 +3212,6 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
 				  EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT);
 	}
 
-	/*
-	 * We may have short-circuited the log tree with the full commit logic
-	 * and left ordered extents on our list, so clear these out to keep us
-	 * from leaking inodes and memory.
-	 */
-	btrfs_free_logged_extents(log, 0);
-	btrfs_free_logged_extents(log, 1);
-
 	free_extent_buffer(log->node);
 	kfree(log);
 }
@@ -3756,7 +3801,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 			       int start_slot, int nr, int inode_only,
 			       u64 logged_isize)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	unsigned long src_offset;
 	unsigned long dst_offset;
 	struct btrfs_root *log = inode->root->log_root;
@@ -3937,9 +3982,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 					struct btrfs_file_extent_item);
 		if (btrfs_file_extent_type(src, extent) ==
 		    BTRFS_FILE_EXTENT_INLINE) {
-			len = btrfs_file_extent_inline_len(src,
-							   src_path->slots[0],
-							   extent);
+			len = btrfs_file_extent_ram_bytes(src, extent);
 			*last_extent = ALIGN(key.offset + len,
 					     fs_info->sectorsize);
 		} else {
@@ -4004,7 +4047,7 @@ fill_holes:
 		extent = btrfs_item_ptr(src, i, struct btrfs_file_extent_item);
 		if (btrfs_file_extent_type(src, extent) ==
 		    BTRFS_FILE_EXTENT_INLINE) {
-			len = btrfs_file_extent_inline_len(src, i, extent);
+			len = btrfs_file_extent_ram_bytes(src, extent);
 			extent_end = ALIGN(key.offset + len,
 					   fs_info->sectorsize);
 		} else {
@@ -4078,131 +4121,32 @@ static int extent_cmp(void *priv, struct list_head *a, struct list_head *b)
 	return 0;
 }
 
-static int wait_ordered_extents(struct btrfs_trans_handle *trans,
-				struct inode *inode,
-				struct btrfs_root *root,
-				const struct extent_map *em,
-				const struct list_head *logged_list,
-				bool *ordered_io_error)
+static int log_extent_csums(struct btrfs_trans_handle *trans,
+			    struct btrfs_inode *inode,
+			    struct btrfs_root *log_root,
+			    const struct extent_map *em)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct btrfs_ordered_extent *ordered;
-	struct btrfs_root *log = root->log_root;
-	u64 mod_start = em->mod_start;
-	u64 mod_len = em->mod_len;
-	const bool skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
 	u64 csum_offset;
 	u64 csum_len;
 	LIST_HEAD(ordered_sums);
 	int ret = 0;
 
-	*ordered_io_error = false;
-
-	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) ||
+	if (inode->flags & BTRFS_INODE_NODATASUM ||
+	    test_bit(EXTENT_FLAG_PREALLOC, &em->flags) ||
 	    em->block_start == EXTENT_MAP_HOLE)
 		return 0;
 
-	/*
-	 * Wait far any ordered extent that covers our extent map. If it
-	 * finishes without an error, first check and see if our csums are on
-	 * our outstanding ordered extents.
-	 */
-	list_for_each_entry(ordered, logged_list, log_list) {
-		struct btrfs_ordered_sum *sum;
-
-		if (!mod_len)
-			break;
-
-		if (ordered->file_offset + ordered->len <= mod_start ||
-		    mod_start + mod_len <= ordered->file_offset)
-			continue;
-
-		if (!test_bit(BTRFS_ORDERED_IO_DONE, &ordered->flags) &&
-		    !test_bit(BTRFS_ORDERED_IOERR, &ordered->flags) &&
-		    !test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags)) {
-			const u64 start = ordered->file_offset;
-			const u64 end = ordered->file_offset + ordered->len - 1;
-
-			WARN_ON(ordered->inode != inode);
-			filemap_fdatawrite_range(inode->i_mapping, start, end);
-		}
-
-		wait_event(ordered->wait,
-			   (test_bit(BTRFS_ORDERED_IO_DONE, &ordered->flags) ||
-			    test_bit(BTRFS_ORDERED_IOERR, &ordered->flags)));
-
-		if (test_bit(BTRFS_ORDERED_IOERR, &ordered->flags)) {
-			/*
-			 * Clear the AS_EIO/AS_ENOSPC flags from the inode's
-			 * i_mapping flags, so that the next fsync won't get
-			 * an outdated io error too.
-			 */
-			filemap_check_errors(inode->i_mapping);
-			*ordered_io_error = true;
-			break;
-		}
-		/*
-		 * We are going to copy all the csums on this ordered extent, so
-		 * go ahead and adjust mod_start and mod_len in case this
-		 * ordered extent has already been logged.
-		 */
-		if (ordered->file_offset > mod_start) {
-			if (ordered->file_offset + ordered->len >=
-			    mod_start + mod_len)
-				mod_len = ordered->file_offset - mod_start;
-			/*
-			 * If we have this case
-			 *
-			 * |--------- logged extent ---------|
-			 *       |----- ordered extent ----|
-			 *
-			 * Just don't mess with mod_start and mod_len, we'll
-			 * just end up logging more csums than we need and it
-			 * will be ok.
-			 */
-		} else {
-			if (ordered->file_offset + ordered->len <
-			    mod_start + mod_len) {
-				mod_len = (mod_start + mod_len) -
-					(ordered->file_offset + ordered->len);
-				mod_start = ordered->file_offset +
-					ordered->len;
-			} else {
-				mod_len = 0;
-			}
-		}
-
-		if (skip_csum)
-			continue;
-
-		/*
-		 * To keep us from looping for the above case of an ordered
-		 * extent that falls inside of the logged extent.
-		 */
-		if (test_and_set_bit(BTRFS_ORDERED_LOGGED_CSUM,
-				     &ordered->flags))
-			continue;
-
-		list_for_each_entry(sum, &ordered->list, list) {
-			ret = btrfs_csum_file_blocks(trans, log, sum);
-			if (ret)
-				break;
-		}
-	}
-
-	if (*ordered_io_error || !mod_len || ret || skip_csum)
-		return ret;
-
+	/* If we're compressed we have to save the entire range of csums. */
 	if (em->compress_type) {
 		csum_offset = 0;
 		csum_len = max(em->block_len, em->orig_block_len);
 	} else {
-		csum_offset = mod_start - em->start;
-		csum_len = mod_len;
+		csum_offset = em->mod_start - em->start;
+		csum_len = em->mod_len;
 	}
 
 	/* block start is already adjusted for the file extent offset. */
-	ret = btrfs_lookup_csums_range(fs_info->csum_root,
+	ret = btrfs_lookup_csums_range(trans->fs_info->csum_root,
 				       em->block_start + csum_offset,
 				       em->block_start + csum_offset +
 				       csum_len - 1, &ordered_sums, 0);
@@ -4214,7 +4158,7 @@ static int wait_ordered_extents(struct btrfs_trans_handle *trans,
 						   struct btrfs_ordered_sum,
 						   list);
 		if (!ret)
-			ret = btrfs_csum_file_blocks(trans, log, sums);
+			ret = btrfs_csum_file_blocks(trans, log_root, sums);
 		list_del(&sums->list);
 		kfree(sums);
 	}
@@ -4226,7 +4170,6 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 			  struct btrfs_inode *inode, struct btrfs_root *root,
 			  const struct extent_map *em,
 			  struct btrfs_path *path,
-			  const struct list_head *logged_list,
 			  struct btrfs_log_ctx *ctx)
 {
 	struct btrfs_root *log = root->log_root;
@@ -4238,18 +4181,11 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 	u64 block_len;
 	int ret;
 	int extent_inserted = 0;
-	bool ordered_io_err = false;
 
-	ret = wait_ordered_extents(trans, &inode->vfs_inode, root, em,
-			logged_list, &ordered_io_err);
+	ret = log_extent_csums(trans, inode, log, em);
 	if (ret)
 		return ret;
 
-	if (ordered_io_err) {
-		ctx->io_err = -EIO;
-		return ctx->io_err;
-	}
-
 	btrfs_init_map_token(&token);
 
 	ret = __btrfs_drop_extents(trans, log, &inode->vfs_inode, path, em->start,
@@ -4424,7 +4360,6 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 				     struct btrfs_root *root,
 				     struct btrfs_inode *inode,
 				     struct btrfs_path *path,
-				     struct list_head *logged_list,
 				     struct btrfs_log_ctx *ctx,
 				     const u64 start,
 				     const u64 end)
@@ -4480,20 +4415,6 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 	}
 
 	list_sort(NULL, &extents, extent_cmp);
-	btrfs_get_logged_extents(inode, logged_list, logged_start, logged_end);
-	/*
-	 * Some ordered extents started by fsync might have completed
-	 * before we could collect them into the list logged_list, which
-	 * means they're gone, not in our logged_list nor in the inode's
-	 * ordered tree. We want the application/user space to know an
-	 * error happened while attempting to persist file data so that
-	 * it can take proper action. If such error happened, we leave
-	 * without writing to the log tree and the fsync must report the
-	 * file data write error and not commit the current transaction.
-	 */
-	ret = filemap_check_errors(inode->vfs_inode.i_mapping);
-	if (ret)
-		ctx->io_err = ret;
 process:
 	while (!list_empty(&extents)) {
 		em = list_entry(extents.next, struct extent_map, list);
@@ -4512,8 +4433,7 @@ process:
 
 		write_unlock(&tree->lock);
 
-		ret = log_one_extent(trans, inode, root, em, path, logged_list,
-				     ctx);
+		ret = log_one_extent(trans, inode, root, em, path, ctx);
 		write_lock(&tree->lock);
 		clear_em_logging(tree, em);
 		free_extent_map(em);
@@ -4712,9 +4632,7 @@ static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans,
 
 		if (btrfs_file_extent_type(leaf, extent) ==
 		    BTRFS_FILE_EXTENT_INLINE) {
-			len = btrfs_file_extent_inline_len(leaf,
-							   path->slots[0],
-							   extent);
+			len = btrfs_file_extent_ram_bytes(leaf, extent);
 			ASSERT(len == i_size ||
 			       (len == fs_info->sectorsize &&
 				btrfs_file_extent_compression(leaf, extent) !=
@@ -4898,7 +4816,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 	struct btrfs_key min_key;
 	struct btrfs_key max_key;
 	struct btrfs_root *log = root->log_root;
-	LIST_HEAD(logged_list);
 	u64 last_extent = 0;
 	int err = 0;
 	int ret;
@@ -5094,8 +5011,7 @@ again:
 				 * we don't need to do more work nor fallback to
 				 * a transaction commit.
 				 */
-				if (IS_ERR(other_inode) &&
-				    PTR_ERR(other_inode) == -ENOENT) {
+				if (other_inode == ERR_PTR(-ENOENT)) {
 					goto next_key;
 				} else if (IS_ERR(other_inode)) {
 					err = PTR_ERR(other_inode);
@@ -5235,7 +5151,7 @@ log_extents:
 	}
 	if (fast_search) {
 		ret = btrfs_log_changed_extents(trans, root, inode, dst_path,
-						&logged_list, ctx, start, end);
+						ctx, start, end);
 		if (ret) {
 			err = ret;
 			goto out_unlock;
@@ -5286,10 +5202,6 @@ log_extents:
 	inode->last_log_commit = inode->last_sub_trans;
 	spin_unlock(&inode->lock);
 out_unlock:
-	if (unlikely(err))
-		btrfs_put_logged_extents(&logged_list);
-	else
-		btrfs_submit_logged_extents(&logged_list, log);
 	mutex_unlock(&inode->log_mutex);
 
 	btrfs_free_path(path);
@@ -5585,7 +5497,7 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 				 struct btrfs_inode *inode,
 				 struct btrfs_log_ctx *ctx)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
 	struct btrfs_path *path;
 	struct btrfs_key key;
@@ -6120,7 +6032,7 @@ int btrfs_log_new_name(struct btrfs_trans_handle *trans,
 			struct btrfs_inode *inode, struct btrfs_inode *old_dir,
 			struct dentry *parent)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 
 	/*
 	 * this will force the logging code to walk the dentry chain

fs/btrfs/volumes.c

@@ -8,15 +8,12 @@
 #include <linux/slab.h>
 #include <linux/buffer_head.h>
 #include <linux/blkdev.h>
-#include <linux/iocontext.h>
-#include <linux/capability.h>
 #include <linux/ratelimit.h>
 #include <linux/kthread.h>
 #include <linux/raid/pq.h>
 #include <linux/semaphore.h>
 #include <linux/uuid.h>
 #include <linux/list_sort.h>
-#include <asm/div64.h>
 #include "ctree.h"
 #include "extent_map.h"
 #include "disk-io.h"
@@ -634,44 +631,48 @@ static void pending_bios_fn(struct btrfs_work *work)
  *		devices.
  */
 static void btrfs_free_stale_devices(const char *path,
-				     struct btrfs_device *skip_dev)
+				     struct btrfs_device *skip_device)
 {
-	struct btrfs_fs_devices *fs_devs, *tmp_fs_devs;
-	struct btrfs_device *dev, *tmp_dev;
+	struct btrfs_fs_devices *fs_devices, *tmp_fs_devices;
+	struct btrfs_device *device, *tmp_device;
 
-	list_for_each_entry_safe(fs_devs, tmp_fs_devs, &fs_uuids, fs_list) {
-
-		if (fs_devs->opened)
+	list_for_each_entry_safe(fs_devices, tmp_fs_devices, &fs_uuids, fs_list) {
+		mutex_lock(&fs_devices->device_list_mutex);
+		if (fs_devices->opened) {
+			mutex_unlock(&fs_devices->device_list_mutex);
 			continue;
+		}
 
-		list_for_each_entry_safe(dev, tmp_dev,
-					 &fs_devs->devices, dev_list) {
+		list_for_each_entry_safe(device, tmp_device,
+					 &fs_devices->devices, dev_list) {
 			int not_found = 0;
 
-			if (skip_dev && skip_dev == dev)
+			if (skip_device && skip_device == device)
 				continue;
-			if (path && !dev->name)
+			if (path && !device->name)
 				continue;
 
 			rcu_read_lock();
 			if (path)
-				not_found = strcmp(rcu_str_deref(dev->name),
+				not_found = strcmp(rcu_str_deref(device->name),
 						   path);
 			rcu_read_unlock();
 			if (not_found)
 				continue;
 
 			/* delete the stale device */
-			if (fs_devs->num_devices == 1) {
-				btrfs_sysfs_remove_fsid(fs_devs);
-				list_del(&fs_devs->fs_list);
-				free_fs_devices(fs_devs);
+			fs_devices->num_devices--;
+			list_del(&device->dev_list);
+			btrfs_free_device(device);
+
+			if (fs_devices->num_devices == 0)
 				break;
-			} else {
-				fs_devs->num_devices--;
-				list_del(&dev->dev_list);
-				btrfs_free_device(dev);
-			}
+		}
+		mutex_unlock(&fs_devices->device_list_mutex);
+		if (fs_devices->num_devices == 0) {
+			btrfs_sysfs_remove_fsid(fs_devices);
+			list_del(&fs_devices->fs_list);
+			free_fs_devices(fs_devices);
 		}
 	}
 }
@@ -750,7 +751,8 @@ error_brelse:
  * error pointer when failed
  */
 static noinline struct btrfs_device *device_list_add(const char *path,
-			   struct btrfs_super_block *disk_super)
+			   struct btrfs_super_block *disk_super,
+			   bool *new_device_added)
 {
 	struct btrfs_device *device;
 	struct btrfs_fs_devices *fs_devices;
@@ -764,21 +766,26 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 		if (IS_ERR(fs_devices))
 			return ERR_CAST(fs_devices);
 
+		mutex_lock(&fs_devices->device_list_mutex);
 		list_add(&fs_devices->fs_list, &fs_uuids);
 
 		device = NULL;
 	} else {
+		mutex_lock(&fs_devices->device_list_mutex);
 		device = find_device(fs_devices, devid,
 				disk_super->dev_item.uuid);
 	}
 
 	if (!device) {
-		if (fs_devices->opened)
+		if (fs_devices->opened) {
+			mutex_unlock(&fs_devices->device_list_mutex);
 			return ERR_PTR(-EBUSY);
+		}
 
 		device = btrfs_alloc_device(NULL, &devid,
 					    disk_super->dev_item.uuid);
 		if (IS_ERR(device)) {
+			mutex_unlock(&fs_devices->device_list_mutex);
 			/* we can safely leave the fs_devices entry around */
 			return device;
 		}
@@ -786,17 +793,16 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 		name = rcu_string_strdup(path, GFP_NOFS);
 		if (!name) {
 			btrfs_free_device(device);
+			mutex_unlock(&fs_devices->device_list_mutex);
 			return ERR_PTR(-ENOMEM);
 		}
 		rcu_assign_pointer(device->name, name);
 
-		mutex_lock(&fs_devices->device_list_mutex);
 		list_add_rcu(&device->dev_list, &fs_devices->devices);
 		fs_devices->num_devices++;
-		mutex_unlock(&fs_devices->device_list_mutex);
 
 		device->fs_devices = fs_devices;
-		btrfs_free_stale_devices(path, device);
+		*new_device_added = true;
 
 		if (disk_super->label[0])
 			pr_info("BTRFS: device label %s devid %llu transid %llu %s\n",
@@ -840,12 +846,15 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 			 * with larger generation number or the last-in if
 			 * generation are equal.
 			 */
+			mutex_unlock(&fs_devices->device_list_mutex);
 			return ERR_PTR(-EEXIST);
 		}
 
 		name = rcu_string_strdup(path, GFP_NOFS);
-		if (!name)
+		if (!name) {
+			mutex_unlock(&fs_devices->device_list_mutex);
 			return ERR_PTR(-ENOMEM);
+		}
 		rcu_string_free(device->name);
 		rcu_assign_pointer(device->name, name);
 		if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) {
@@ -865,6 +874,7 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 
 	fs_devices->total_devices = btrfs_super_num_devices(disk_super);
 
+	mutex_unlock(&fs_devices->device_list_mutex);
 	return device;
 }
 
@@ -1004,7 +1014,7 @@ static void btrfs_close_bdev(struct btrfs_device *device)
 	blkdev_put(device->bdev, device->mode);
 }
 
-static void btrfs_prepare_close_one_device(struct btrfs_device *device)
+static void btrfs_close_one_device(struct btrfs_device *device)
 {
 	struct btrfs_fs_devices *fs_devices = device->fs_devices;
 	struct btrfs_device *new_device;
@@ -1022,6 +1032,8 @@ static void btrfs_prepare_close_one_device(struct btrfs_device *device)
 	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
 		fs_devices->missing_devices--;
 
+	btrfs_close_bdev(device);
+
 	new_device = btrfs_alloc_device(NULL, &device->devid,
 					device->uuid);
 	BUG_ON(IS_ERR(new_device)); /* -ENOMEM */
@@ -1035,39 +1047,23 @@ static void btrfs_prepare_close_one_device(struct btrfs_device *device)
 
 	list_replace_rcu(&device->dev_list, &new_device->dev_list);
 	new_device->fs_devices = device->fs_devices;
+
+	call_rcu(&device->rcu, free_device_rcu);
 }
 
 static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
 {
 	struct btrfs_device *device, *tmp;
-	struct list_head pending_put;
-
-	INIT_LIST_HEAD(&pending_put);
 
 	if (--fs_devices->opened > 0)
 		return 0;
 
 	mutex_lock(&fs_devices->device_list_mutex);
 	list_for_each_entry_safe(device, tmp, &fs_devices->devices, dev_list) {
-		btrfs_prepare_close_one_device(device);
-		list_add(&device->dev_list, &pending_put);
+		btrfs_close_one_device(device);
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
 
-	/*
-	 * btrfs_show_devname() is using the device_list_mutex,
-	 * sometimes call to blkdev_put() leads vfs calling
-	 * into this func. So do put outside of device_list_mutex,
-	 * as of now.
-	 */
-	while (!list_empty(&pending_put)) {
-		device = list_first_entry(&pending_put,
-				struct btrfs_device, dev_list);
-		list_del(&device->dev_list);
-		btrfs_close_bdev(device);
-		call_rcu(&device->rcu, free_device_rcu);
-	}
-
 	WARN_ON(fs_devices->open_devices);
 	WARN_ON(fs_devices->rw_devices);
 	fs_devices->opened = 0;
@@ -1146,7 +1142,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 {
 	int ret;
 
-	mutex_lock(&uuid_mutex);
+	lockdep_assert_held(&uuid_mutex);
+
 	mutex_lock(&fs_devices->device_list_mutex);
 	if (fs_devices->opened) {
 		fs_devices->opened++;
@@ -1156,7 +1153,6 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		ret = open_fs_devices(fs_devices, flags, holder);
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
-	mutex_unlock(&uuid_mutex);
 
 	return ret;
 }
@@ -1217,16 +1213,18 @@ static int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
  * and we are not allowed to call set_blocksize during the scan. The superblock
  * is read via pagecache
  */
-int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
-			  struct btrfs_fs_devices **fs_devices_ret)
+struct btrfs_device *btrfs_scan_one_device(const char *path, fmode_t flags,
+					   void *holder)
 {
 	struct btrfs_super_block *disk_super;
-	struct btrfs_device *device;
+	bool new_device_added = false;
+	struct btrfs_device *device = NULL;
 	struct block_device *bdev;
 	struct page *page;
-	int ret = 0;
 	u64 bytenr;
 
+	lockdep_assert_held(&uuid_mutex);
+
 	/*
 	 * we would like to check all the supers, but that would make
 	 * a btrfs mount succeed after a mkfs from a different FS.
@@ -1238,112 +1236,25 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 
 	bdev = blkdev_get_by_path(path, flags, holder);
 	if (IS_ERR(bdev))
-		return PTR_ERR(bdev);
+		return ERR_CAST(bdev);
 
 	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super)) {
-		ret = -EINVAL;
+		device = ERR_PTR(-EINVAL);
 		goto error_bdev_put;
 	}
 
-	mutex_lock(&uuid_mutex);
-	device = device_list_add(path, disk_super);
-	if (IS_ERR(device))
-		ret = PTR_ERR(device);
-	else
-		*fs_devices_ret = device->fs_devices;
-	mutex_unlock(&uuid_mutex);
+	device = device_list_add(path, disk_super, &new_device_added);
+	if (!IS_ERR(device)) {
+		if (new_device_added)
+			btrfs_free_stale_devices(path, device);
+	}
 
 	btrfs_release_disk_super(page);
 
 error_bdev_put:
 	blkdev_put(bdev, flags);
 
-	return ret;
-}
-
-/* helper to account the used device space in the range */
-int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
-				   u64 end, u64 *length)
-{
-	struct btrfs_key key;
-	struct btrfs_root *root = device->fs_info->dev_root;
-	struct btrfs_dev_extent *dev_extent;
-	struct btrfs_path *path;
-	u64 extent_end;
-	int ret;
-	int slot;
-	struct extent_buffer *l;
-
-	*length = 0;
-
-	if (start >= device->total_bytes ||
-		test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state))
-		return 0;
-
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-	path->reada = READA_FORWARD;
-
-	key.objectid = device->devid;
-	key.offset = start;
-	key.type = BTRFS_DEV_EXTENT_KEY;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-	if (ret > 0) {
-		ret = btrfs_previous_item(root, path, key.objectid, key.type);
-		if (ret < 0)
-			goto out;
-	}
-
-	while (1) {
-		l = path->nodes[0];
-		slot = path->slots[0];
-		if (slot >= btrfs_header_nritems(l)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret == 0)
-				continue;
-			if (ret < 0)
-				goto out;
-
-			break;
-		}
-		btrfs_item_key_to_cpu(l, &key, slot);
-
-		if (key.objectid < device->devid)
-			goto next;
-
-		if (key.objectid > device->devid)
-			break;
-
-		if (key.type != BTRFS_DEV_EXTENT_KEY)
-			goto next;
-
-		dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent);
-		extent_end = key.offset + btrfs_dev_extent_length(l,
-								  dev_extent);
-		if (key.offset <= start && extent_end > end) {
-			*length = end - start + 1;
-			break;
-		} else if (key.offset <= start && extent_end > start)
-			*length += extent_end - start;
-		else if (key.offset > start && extent_end <= end)
-			*length += extent_end - key.offset;
-		else if (key.offset > start && key.offset <= end) {
-			*length += end - key.offset + 1;
-			break;
-		} else if (key.offset > end)
-			break;
-
-next:
-		path->slots[0]++;
-	}
-	ret = 0;
-out:
-	btrfs_free_path(path);
-	return ret;
+	return device;
 }
 
 static int contains_pending_extent(struct btrfs_transaction *transaction,
@@ -1755,10 +1666,8 @@ error:
  * the btrfs_device struct should be fully filled in
  */
 static int btrfs_add_dev_item(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info,
 			    struct btrfs_device *device)
 {
-	struct btrfs_root *root = fs_info->chunk_root;
 	int ret;
 	struct btrfs_path *path;
 	struct btrfs_dev_item *dev_item;
@@ -1774,8 +1683,8 @@ static int btrfs_add_dev_item(struct btrfs_trans_handle *trans,
 	key.type = BTRFS_DEV_ITEM_KEY;
 	key.offset = device->devid;
 
-	ret = btrfs_insert_empty_item(trans, root, path, &key,
-				      sizeof(*dev_item));
+	ret = btrfs_insert_empty_item(trans, trans->fs_info->chunk_root, path,
+				      &key, sizeof(*dev_item));
 	if (ret)
 		goto out;
 
@@ -1800,7 +1709,7 @@ static int btrfs_add_dev_item(struct btrfs_trans_handle *trans,
 	ptr = btrfs_device_uuid(dev_item);
 	write_extent_buffer(leaf, device->uuid, ptr, BTRFS_UUID_SIZE);
 	ptr = btrfs_device_fsid(dev_item);
-	write_extent_buffer(leaf, fs_info->fsid, ptr, BTRFS_FSID_SIZE);
+	write_extent_buffer(leaf, trans->fs_info->fsid, ptr, BTRFS_FSID_SIZE);
 	btrfs_mark_buffer_dirty(leaf);
 
 	ret = 0;
@@ -1924,9 +1833,10 @@ static struct btrfs_device * btrfs_find_next_active_device(
  * where this function called, there should be always be another device (or
  * this_dev) which is active.
  */
-void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
-		struct btrfs_device *device, struct btrfs_device *this_dev)
+void btrfs_assign_next_active_device(struct btrfs_device *device,
+				     struct btrfs_device *this_dev)
 {
+	struct btrfs_fs_info *fs_info = device->fs_info;
 	struct btrfs_device *next_device;
 
 	if (this_dev)
@@ -2029,11 +1939,14 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 
 	cur_devices->num_devices--;
 	cur_devices->total_devices--;
+	/* Update total_devices of the parent fs_devices if it's seed */
+	if (cur_devices != fs_devices)
+		fs_devices->total_devices--;
 
 	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
 		cur_devices->missing_devices--;
 
-	btrfs_assign_next_active_device(fs_info, device, NULL);
+	btrfs_assign_next_active_device(device, NULL);
 
 	if (device->bdev) {
 		cur_devices->open_devices--;
@@ -2084,12 +1997,11 @@ error_undo:
 	goto out;
 }
 
-void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
-					struct btrfs_device *srcdev)
+void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev)
 {
 	struct btrfs_fs_devices *fs_devices;
 
-	lockdep_assert_held(&fs_info->fs_devices->device_list_mutex);
+	lockdep_assert_held(&srcdev->fs_info->fs_devices->device_list_mutex);
 
 	/*
 	 * in case of fs with no seed, srcdev->fs_devices will point
@@ -2151,10 +2063,9 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
 	}
 }
 
-void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
-				      struct btrfs_device *tgtdev)
+void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev)
 {
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	struct btrfs_fs_devices *fs_devices = tgtdev->fs_info->fs_devices;
 
 	WARN_ON(!tgtdev);
 	mutex_lock(&fs_devices->device_list_mutex);
@@ -2166,7 +2077,7 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 
 	fs_devices->num_devices--;
 
-	btrfs_assign_next_active_device(fs_info, tgtdev, NULL);
+	btrfs_assign_next_active_device(tgtdev, NULL);
 
 	list_del_rcu(&tgtdev->dev_list);
 
@@ -2297,7 +2208,7 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 	INIT_LIST_HEAD(&seed_devices->alloc_list);
 	mutex_init(&seed_devices->device_list_mutex);
 
-	mutex_lock(&fs_info->fs_devices->device_list_mutex);
+	mutex_lock(&fs_devices->device_list_mutex);
 	list_splice_init_rcu(&fs_devices->devices, &seed_devices->devices,
 			      synchronize_rcu);
 	list_for_each_entry(device, &seed_devices->devices, dev_list)
@@ -2317,7 +2228,7 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 	generate_random_uuid(fs_devices->fsid);
 	memcpy(fs_info->fsid, fs_devices->fsid, BTRFS_FSID_SIZE);
 	memcpy(disk_super->fsid, fs_devices->fsid, BTRFS_FSID_SIZE);
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
 
 	super_flags = btrfs_super_flags(disk_super) &
 		      ~BTRFS_SUPER_FLAG_SEEDING;
@@ -2407,15 +2318,16 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	struct btrfs_trans_handle *trans;
 	struct btrfs_device *device;
 	struct block_device *bdev;
-	struct list_head *devices;
 	struct super_block *sb = fs_info->sb;
 	struct rcu_string *name;
-	u64 tmp;
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	u64 orig_super_total_bytes;
+	u64 orig_super_num_devices;
 	int seeding_dev = 0;
 	int ret = 0;
 	bool unlocked = false;
 
-	if (sb_rdonly(sb) && !fs_info->fs_devices->seeding)
+	if (sb_rdonly(sb) && !fs_devices->seeding)
 		return -EROFS;
 
 	bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
@@ -2423,7 +2335,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	if (IS_ERR(bdev))
 		return PTR_ERR(bdev);
 
-	if (fs_info->fs_devices->seeding) {
+	if (fs_devices->seeding) {
 		seeding_dev = 1;
 		down_write(&sb->s_umount);
 		mutex_lock(&uuid_mutex);
@@ -2431,18 +2343,16 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 
 	filemap_write_and_wait(bdev->bd_inode->i_mapping);
 
-	devices = &fs_info->fs_devices->devices;
-
-	mutex_lock(&fs_info->fs_devices->device_list_mutex);
-	list_for_each_entry(device, devices, dev_list) {
+	mutex_lock(&fs_devices->device_list_mutex);
+	list_for_each_entry(device, &fs_devices->devices, dev_list) {
 		if (device->bdev == bdev) {
 			ret = -EEXIST;
 			mutex_unlock(
-				&fs_info->fs_devices->device_list_mutex);
+				&fs_devices->device_list_mutex);
 			goto error;
 		}
 	}
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
 
 	device = btrfs_alloc_device(fs_info, NULL, NULL);
 	if (IS_ERR(device)) {
@@ -2491,33 +2401,34 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		}
 	}
 
-	device->fs_devices = fs_info->fs_devices;
+	device->fs_devices = fs_devices;
 
-	mutex_lock(&fs_info->fs_devices->device_list_mutex);
+	mutex_lock(&fs_devices->device_list_mutex);
 	mutex_lock(&fs_info->chunk_mutex);
-	list_add_rcu(&device->dev_list, &fs_info->fs_devices->devices);
-	list_add(&device->dev_alloc_list,
-		 &fs_info->fs_devices->alloc_list);
-	fs_info->fs_devices->num_devices++;
-	fs_info->fs_devices->open_devices++;
-	fs_info->fs_devices->rw_devices++;
-	fs_info->fs_devices->total_devices++;
-	fs_info->fs_devices->total_rw_bytes += device->total_bytes;
+	list_add_rcu(&device->dev_list, &fs_devices->devices);
+	list_add(&device->dev_alloc_list, &fs_devices->alloc_list);
+	fs_devices->num_devices++;
+	fs_devices->open_devices++;
+	fs_devices->rw_devices++;
+	fs_devices->total_devices++;
+	fs_devices->total_rw_bytes += device->total_bytes;
 
 	atomic64_add(device->total_bytes, &fs_info->free_chunk_space);
 
 	if (!blk_queue_nonrot(q))
-		fs_info->fs_devices->rotating = 1;
+		fs_devices->rotating = 1;
 
-	tmp = btrfs_super_total_bytes(fs_info->super_copy);
+	orig_super_total_bytes = btrfs_super_total_bytes(fs_info->super_copy);
 	btrfs_set_super_total_bytes(fs_info->super_copy,
-		round_down(tmp + device->total_bytes, fs_info->sectorsize));
+		round_down(orig_super_total_bytes + device->total_bytes,
+			   fs_info->sectorsize));
 
-	tmp = btrfs_super_num_devices(fs_info->super_copy);
-	btrfs_set_super_num_devices(fs_info->super_copy, tmp + 1);
+	orig_super_num_devices = btrfs_super_num_devices(fs_info->super_copy);
+	btrfs_set_super_num_devices(fs_info->super_copy,
+				    orig_super_num_devices + 1);
 
 	/* add sysfs device entry */
-	btrfs_sysfs_add_device_link(fs_info->fs_devices, device);
+	btrfs_sysfs_add_device_link(fs_devices, device);
 
 	/*
 	 * we've got more storage, clear any full flags on the space
@@ -2526,7 +2437,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	btrfs_clear_space_info_full(fs_info);
 
 	mutex_unlock(&fs_info->chunk_mutex);
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
 
 	if (seeding_dev) {
 		mutex_lock(&fs_info->chunk_mutex);
@@ -2538,7 +2449,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		}
 	}
 
-	ret = btrfs_add_dev_item(trans, fs_info, device);
+	ret = btrfs_add_dev_item(trans, device);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto error_sysfs;
@@ -2558,7 +2469,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		 */
 		snprintf(fsid_buf, BTRFS_UUID_UNPARSED_SIZE, "%pU",
 						fs_info->fsid);
-		if (kobject_rename(&fs_info->fs_devices->fsid_kobj, fsid_buf))
+		if (kobject_rename(&fs_devices->fsid_kobj, fsid_buf))
 			btrfs_warn(fs_info,
 				   "sysfs: failed to create fsid for sprout");
 	}
@@ -2593,7 +2504,23 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	return ret;
 
 error_sysfs:
-	btrfs_sysfs_rm_device_link(fs_info->fs_devices, device);
+	btrfs_sysfs_rm_device_link(fs_devices, device);
+	mutex_lock(&fs_info->fs_devices->device_list_mutex);
+	mutex_lock(&fs_info->chunk_mutex);
+	list_del_rcu(&device->dev_list);
+	list_del(&device->dev_alloc_list);
+	fs_info->fs_devices->num_devices--;
+	fs_info->fs_devices->open_devices--;
+	fs_info->fs_devices->rw_devices--;
+	fs_info->fs_devices->total_devices--;
+	fs_info->fs_devices->total_rw_bytes -= device->total_bytes;
+	atomic64_sub(device->total_bytes, &fs_info->free_chunk_space);
+	btrfs_set_super_total_bytes(fs_info->super_copy,
+				    orig_super_total_bytes);
+	btrfs_set_super_num_devices(fs_info->super_copy,
+				    orig_super_num_devices);
+	mutex_unlock(&fs_info->chunk_mutex);
+	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 error_trans:
 	if (seeding_dev)
 		sb->s_flags |= SB_RDONLY;
@@ -2697,9 +2624,9 @@ int btrfs_grow_device(struct btrfs_trans_handle *trans,
 	return btrfs_update_device(trans, device);
 }
 
-static int btrfs_free_chunk(struct btrfs_trans_handle *trans,
-			    struct btrfs_fs_info *fs_info, u64 chunk_offset)
+static int btrfs_free_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = fs_info->chunk_root;
 	int ret;
 	struct btrfs_path *path;
@@ -2808,9 +2735,9 @@ static struct extent_map *get_chunk_map(struct btrfs_fs_info *fs_info,
 	return em;
 }
 
-int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info, u64 chunk_offset)
+int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct extent_map *em;
 	struct map_lookup *map;
 	u64 dev_extent_len = 0;
@@ -2829,7 +2756,7 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 	}
 	map = em->map_lookup;
 	mutex_lock(&fs_info->chunk_mutex);
-	check_system_chunk(trans, fs_info, map->type);
+	check_system_chunk(trans, map->type);
 	mutex_unlock(&fs_info->chunk_mutex);
 
 	/*
@@ -2869,7 +2796,7 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
 
-	ret = btrfs_free_chunk(trans, fs_info, chunk_offset);
+	ret = btrfs_free_chunk(trans, chunk_offset);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
@@ -2885,7 +2812,7 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 		}
 	}
 
-	ret = btrfs_remove_block_group(trans, fs_info, chunk_offset, em);
+	ret = btrfs_remove_block_group(trans, chunk_offset, em);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
@@ -2950,7 +2877,7 @@ static int btrfs_relocate_chunk(struct btrfs_fs_info *fs_info, u64 chunk_offset)
 	 * step two, delete the device extents and the
 	 * chunk tree entries
 	 */
-	ret = btrfs_remove_chunk(trans, fs_info, chunk_offset);
+	ret = btrfs_remove_chunk(trans, chunk_offset);
 	btrfs_end_transaction(trans);
 	return ret;
 }
@@ -3059,7 +2986,7 @@ static int btrfs_may_alloc_data_chunk(struct btrfs_fs_info *fs_info,
 			if (IS_ERR(trans))
 				return PTR_ERR(trans);
 
-			ret = btrfs_force_chunk_alloc(trans, fs_info,
+			ret = btrfs_force_chunk_alloc(trans,
 						      BTRFS_BLOCK_GROUP_DATA);
 			btrfs_end_transaction(trans);
 			if (ret < 0)
@@ -4692,7 +4619,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 
 	if (type & BTRFS_BLOCK_GROUP_DATA) {
 		max_stripe_size = SZ_1G;
-		max_chunk_size = 10 * max_stripe_size;
+		max_chunk_size = BTRFS_MAX_DATA_CHUNK_SIZE;
 		if (!devs_max)
 			devs_max = BTRFS_MAX_DEVS(info);
 	} else if (type & BTRFS_BLOCK_GROUP_METADATA) {
@@ -4900,7 +4827,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	refcount_inc(&em->refs);
 	write_unlock(&em_tree->lock);
 
-	ret = btrfs_make_block_group(trans, info, 0, type, start, num_bytes);
+	ret = btrfs_make_block_group(trans, 0, type, start, num_bytes);
 	if (ret)
 		goto error_del_extent;
 
@@ -4934,9 +4861,9 @@ error:
 }
 
 int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
-				u64 chunk_offset, u64 chunk_size)
+			     u64 chunk_offset, u64 chunk_size)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	struct btrfs_root *chunk_root = fs_info->chunk_root;
 	struct btrfs_key key;
@@ -5038,13 +4965,12 @@ out:
  * require modifying the chunk tree. This division is important for the
  * bootstrap process of adding storage to a seed btrfs.
  */
-int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info, u64 type)
+int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type)
 {
 	u64 chunk_offset;
 
-	lockdep_assert_held(&fs_info->chunk_mutex);
-	chunk_offset = find_next_chunk(fs_info);
+	lockdep_assert_held(&trans->fs_info->chunk_mutex);
+	chunk_offset = find_next_chunk(trans->fs_info);
 	return __btrfs_alloc_chunk(trans, chunk_offset, type);
 }
 
@@ -5175,7 +5101,7 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
 		/*
 		 * There could be two corrupted data stripes, we need
 		 * to loop retry in order to rebuild the correct data.
-		 * 
+		 *
 		 * Fail a stripe at a time on every retry except the
 		 * stripe under reconstruction.
 		 */
@@ -6187,21 +6113,11 @@ static void submit_stripe_bio(struct btrfs_bio *bbio, struct bio *bio,
 	btrfs_io_bio(bio)->stripe_index = dev_nr;
 	bio->bi_end_io = btrfs_end_bio;
 	bio->bi_iter.bi_sector = physical >> 9;
-#ifdef DEBUG
-	{
-		struct rcu_string *name;
-
-		rcu_read_lock();
-		name = rcu_dereference(dev->name);
-		btrfs_debug(fs_info,
-			"btrfs_map_bio: rw %d 0x%x, sector=%llu, dev=%lu (%s id %llu), size=%u",
-			bio_op(bio), bio->bi_opf,
-			(u64)bio->bi_iter.bi_sector,
-			(u_long)dev->bdev->bd_dev, name->str, dev->devid,
-			bio->bi_iter.bi_size);
-		rcu_read_unlock();
-	}
-#endif
+	btrfs_debug_in_rcu(fs_info,
+	"btrfs_map_bio: rw %d 0x%x, sector=%llu, dev=%lu (%s id %llu), size=%u",
+		bio_op(bio), bio->bi_opf, (u64)bio->bi_iter.bi_sector,
+		(u_long)dev->bdev->bd_dev, rcu_str_deref(dev->name), dev->devid,
+		bio->bi_iter.bi_size);
 	bio_set_dev(bio, dev->bdev);
 
 	btrfs_bio_counter_inc_noblocked(fs_info);
@@ -6403,6 +6319,8 @@ static int btrfs_check_chunk_valid(struct btrfs_fs_info *fs_info,
 	u16 num_stripes;
 	u16 sub_stripes;
 	u64 type;
+	u64 features;
+	bool mixed = false;
 
 	length = btrfs_chunk_length(leaf, chunk);
 	stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
@@ -6441,6 +6359,32 @@ static int btrfs_check_chunk_valid(struct btrfs_fs_info *fs_info,
 			  btrfs_chunk_type(leaf, chunk));
 		return -EIO;
 	}
+
+	if ((type & BTRFS_BLOCK_GROUP_TYPE_MASK) == 0) {
+		btrfs_err(fs_info, "missing chunk type flag: 0x%llx", type);
+		return -EIO;
+	}
+
+	if ((type & BTRFS_BLOCK_GROUP_SYSTEM) &&
+	    (type & (BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA))) {
+		btrfs_err(fs_info,
+			"system chunk with data or metadata type: 0x%llx", type);
+		return -EIO;
+	}
+
+	features = btrfs_super_incompat_flags(fs_info->super_copy);
+	if (features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS)
+		mixed = true;
+
+	if (!mixed) {
+		if ((type & BTRFS_BLOCK_GROUP_METADATA) &&
+		    (type & BTRFS_BLOCK_GROUP_DATA)) {
+			btrfs_err(fs_info,
+			"mixed chunk type in non-mixed mode: 0x%llx", type);
+			return -EIO;
+		}
+	}
+
 	if ((type & BTRFS_BLOCK_GROUP_RAID10 && sub_stripes != 2) ||
 	    (type & BTRFS_BLOCK_GROUP_RAID1 && num_stripes < 1) ||
 	    (type & BTRFS_BLOCK_GROUP_RAID5 && num_stripes < 2) ||
@@ -6527,6 +6471,7 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 	map->stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
 	map->type = btrfs_chunk_type(leaf, chunk);
 	map->sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk);
+	map->verified_stripes = 0;
 	for (i = 0; i < num_stripes; i++) {
 		map->stripes[i].physical =
 			btrfs_stripe_offset_nr(leaf, chunk, i);
@@ -6563,10 +6508,14 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 	write_lock(&map_tree->map_tree.lock);
 	ret = add_extent_mapping(&map_tree->map_tree, em, 0);
 	write_unlock(&map_tree->map_tree.lock);
-	BUG_ON(ret); /* Tree corruption */
+	if (ret < 0) {
+		btrfs_err(fs_info,
+			  "failed to add chunk map, start=%llu len=%llu: %d",
+			  em->start, em->len, ret);
+	}
 	free_extent_map(em);
 
-	return 0;
+	return ret;
 }
 
 static void fill_device_from_item(struct extent_buffer *leaf,
@@ -7108,9 +7057,9 @@ out:
 }
 
 static int update_dev_stat_item(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
 				struct btrfs_device *device)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *dev_root = fs_info->dev_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
@@ -7203,7 +7152,7 @@ int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
 		 */
 		smp_rmb();
 
-		ret = update_dev_stat_item(trans, fs_info, device);
+		ret = update_dev_stat_item(trans, device);
 		if (!ret)
 			atomic_sub(stats_cnt, &device->dev_stats_ccnt);
 	}
@@ -7382,3 +7331,197 @@ void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info)
 		fs_devices = fs_devices->seed;
 	}
 }
+
+/*
+ * Multiplicity factor for simple profiles: DUP, RAID1-like and RAID10.
+ */
+int btrfs_bg_type_to_factor(u64 flags)
+{
+	if (flags & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1 |
+		     BTRFS_BLOCK_GROUP_RAID10))
+		return 2;
+	return 1;
+}
+
+
+static u64 calc_stripe_length(u64 type, u64 chunk_len, int num_stripes)
+{
+	int index = btrfs_bg_flags_to_raid_index(type);
+	int ncopies = btrfs_raid_array[index].ncopies;
+	int data_stripes;
+
+	switch (type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
+	case BTRFS_BLOCK_GROUP_RAID5:
+		data_stripes = num_stripes - 1;
+		break;
+	case BTRFS_BLOCK_GROUP_RAID6:
+		data_stripes = num_stripes - 2;
+		break;
+	default:
+		data_stripes = num_stripes / ncopies;
+		break;
+	}
+	return div_u64(chunk_len, data_stripes);
+}
+
+static int verify_one_dev_extent(struct btrfs_fs_info *fs_info,
+				 u64 chunk_offset, u64 devid,
+				 u64 physical_offset, u64 physical_len)
+{
+	struct extent_map_tree *em_tree = &fs_info->mapping_tree.map_tree;
+	struct extent_map *em;
+	struct map_lookup *map;
+	u64 stripe_len;
+	bool found = false;
+	int ret = 0;
+	int i;
+
+	read_lock(&em_tree->lock);
+	em = lookup_extent_mapping(em_tree, chunk_offset, 1);
+	read_unlock(&em_tree->lock);
+
+	if (!em) {
+		btrfs_err(fs_info,
+"dev extent physical offset %llu on devid %llu doesn't have corresponding chunk",
+			  physical_offset, devid);
+		ret = -EUCLEAN;
+		goto out;
+	}
+
+	map = em->map_lookup;
+	stripe_len = calc_stripe_length(map->type, em->len, map->num_stripes);
+	if (physical_len != stripe_len) {
+		btrfs_err(fs_info,
+"dev extent physical offset %llu on devid %llu length doesn't match chunk %llu, have %llu expect %llu",
+			  physical_offset, devid, em->start, physical_len,
+			  stripe_len);
+		ret = -EUCLEAN;
+		goto out;
+	}
+
+	for (i = 0; i < map->num_stripes; i++) {
+		if (map->stripes[i].dev->devid == devid &&
+		    map->stripes[i].physical == physical_offset) {
+			found = true;
+			if (map->verified_stripes >= map->num_stripes) {
+				btrfs_err(fs_info,
+				"too many dev extents for chunk %llu found",
+					  em->start);
+				ret = -EUCLEAN;
+				goto out;
+			}
+			map->verified_stripes++;
+			break;
+		}
+	}
+	if (!found) {
+		btrfs_err(fs_info,
+	"dev extent physical offset %llu devid %llu has no corresponding chunk",
+			physical_offset, devid);
+		ret = -EUCLEAN;
+	}
+out:
+	free_extent_map(em);
+	return ret;
+}
+
+static int verify_chunk_dev_extent_mapping(struct btrfs_fs_info *fs_info)
+{
+	struct extent_map_tree *em_tree = &fs_info->mapping_tree.map_tree;
+	struct extent_map *em;
+	struct rb_node *node;
+	int ret = 0;
+
+	read_lock(&em_tree->lock);
+	for (node = rb_first(&em_tree->map); node; node = rb_next(node)) {
+		em = rb_entry(node, struct extent_map, rb_node);
+		if (em->map_lookup->num_stripes !=
+		    em->map_lookup->verified_stripes) {
+			btrfs_err(fs_info,
+			"chunk %llu has missing dev extent, have %d expect %d",
+				  em->start, em->map_lookup->verified_stripes,
+				  em->map_lookup->num_stripes);
+			ret = -EUCLEAN;
+			goto out;
+		}
+	}
+out:
+	read_unlock(&em_tree->lock);
+	return ret;
+}
+
+/*
+ * Ensure that all dev extents are mapped to correct chunk, otherwise
+ * later chunk allocation/free would cause unexpected behavior.
+ *
+ * NOTE: This will iterate through the whole device tree, which should be of
+ * the same size level as the chunk tree.  This slightly increases mount time.
+ */
+int btrfs_verify_dev_extents(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_path *path;
+	struct btrfs_root *root = fs_info->dev_root;
+	struct btrfs_key key;
+	int ret = 0;
+
+	key.objectid = 1;
+	key.type = BTRFS_DEV_EXTENT_KEY;
+	key.offset = 0;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	path->reada = READA_FORWARD;
+	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0)
+		goto out;
+
+	if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
+		ret = btrfs_next_item(root, path);
+		if (ret < 0)
+			goto out;
+		/* No dev extents at all? Not good */
+		if (ret > 0) {
+			ret = -EUCLEAN;
+			goto out;
+		}
+	}
+	while (1) {
+		struct extent_buffer *leaf = path->nodes[0];
+		struct btrfs_dev_extent *dext;
+		int slot = path->slots[0];
+		u64 chunk_offset;
+		u64 physical_offset;
+		u64 physical_len;
+		u64 devid;
+
+		btrfs_item_key_to_cpu(leaf, &key, slot);
+		if (key.type != BTRFS_DEV_EXTENT_KEY)
+			break;
+		devid = key.objectid;
+		physical_offset = key.offset;
+
+		dext = btrfs_item_ptr(leaf, slot, struct btrfs_dev_extent);
+		chunk_offset = btrfs_dev_extent_chunk_offset(leaf, dext);
+		physical_len = btrfs_dev_extent_length(leaf, dext);
+
+		ret = verify_one_dev_extent(fs_info, chunk_offset, devid,
+					    physical_offset, physical_len);
+		if (ret < 0)
+			goto out;
+		ret = btrfs_next_item(root, path);
+		if (ret < 0)
+			goto out;
+		if (ret > 0) {
+			ret = 0;
+			break;
+		}
+	}
+
+	/* Ensure all chunks have corresponding dev extents */
+	ret = verify_chunk_dev_extent_mapping(fs_info);
+out:
+	btrfs_free_path(path);
+	return ret;
+}

fs/btrfs/volumes.h

@@ -11,6 +11,8 @@
 #include <linux/btrfs.h>
 #include "async-thread.h"
 
+#define BTRFS_MAX_DATA_CHUNK_SIZE	(10ULL * SZ_1G)
+
 extern struct mutex uuid_mutex;
 
 #define BTRFS_STRIPE_LEN	SZ_64K
@@ -343,6 +345,7 @@ struct map_lookup {
 	u64 stripe_len;
 	int num_stripes;
 	int sub_stripes;
+	int verified_stripes; /* For mount time dev extent verification */
 	struct btrfs_bio_stripe stripes[];
 };
 
@@ -382,8 +385,6 @@ static inline enum btrfs_map_op btrfs_op(struct bio *bio)
 	}
 }
 
-int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
-				   u64 end, u64 *length);
 void btrfs_get_bbio(struct btrfs_bio *bbio);
 void btrfs_put_bbio(struct btrfs_bio *bbio);
 int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
@@ -396,20 +397,19 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
 		     u64 physical, u64 **logical, int *naddrs, int *stripe_len);
 int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
 int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
-int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info, u64 type);
+int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type);
 void btrfs_mapping_init(struct btrfs_mapping_tree *tree);
 void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree);
 blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 			   int mirror_num, int async_submit);
 int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       fmode_t flags, void *holder);
-int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
-			  struct btrfs_fs_devices **fs_devices_ret);
+struct btrfs_device *btrfs_scan_one_device(const char *path,
+					   fmode_t flags, void *holder);
 int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
 void btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, int step);
-void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
-		struct btrfs_device *device, struct btrfs_device *this_dev);
+void btrfs_assign_next_active_device(struct btrfs_device *device,
+				     struct btrfs_device *this_dev);
 int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
 					 const char *device_path,
 					 struct btrfs_device **device);
@@ -453,22 +453,18 @@ void btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
 int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info);
 int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
 			struct btrfs_fs_info *fs_info);
-void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
-					struct btrfs_device *srcdev);
+void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev);
 void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
 				      struct btrfs_device *srcdev);
-void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
-				      struct btrfs_device *tgtdev);
+void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev);
 void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path);
 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);
 int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
-				u64 chunk_offset, u64 chunk_size);
-int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
-		       struct btrfs_fs_info *fs_info, u64 chunk_offset);
+			     u64 chunk_offset, u64 chunk_size);
+int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset);
 
 static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
 				      int index)
@@ -560,4 +556,7 @@ void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
 bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info,
 					struct btrfs_device *failing_dev);
 
+int btrfs_bg_type_to_factor(u64 flags);
+int btrfs_verify_dev_extents(struct btrfs_fs_info *fs_info);
+
 #endif

include/trace/events/btrfs.h

@@ -374,7 +374,7 @@ DECLARE_EVENT_CLASS(
 		__entry->extent_type	= btrfs_file_extent_type(l, fi);
 		__entry->compression	= btrfs_file_extent_compression(l, fi);
 		__entry->extent_start	= start;
-		__entry->extent_end	= (start + btrfs_file_extent_inline_len(l, slot, fi));
+		__entry->extent_end	= (start + btrfs_file_extent_ram_bytes(l, fi));
 	),
 
 	TP_printk_btrfs(
@@ -433,7 +433,6 @@ DEFINE_EVENT(
 		{ (1 << BTRFS_ORDERED_DIRECT),	 	"DIRECT" 	}, \
 		{ (1 << BTRFS_ORDERED_IOERR), 		"IOERR" 	}, \
 		{ (1 << BTRFS_ORDERED_UPDATED_ISIZE), 	"UPDATED_ISIZE"	}, \
-		{ (1 << BTRFS_ORDERED_LOGGED_CSUM), 	"LOGGED_CSUM"	}, \
 		{ (1 << BTRFS_ORDERED_TRUNCATED), 	"TRUNCATED"	})