Merge tag 'for-5.6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "Features, highlights:

   - async discard
       - "mount -o discard=async" to enable it
       - freed extents are not discarded immediatelly, but grouped
         together and trimmed later, with IO rate limiting
       - the "sync" mode submits short extents that could have been
         ignored completely by the device, for SATA prior to 3.1 the
         requests are unqueued and have a big impact on performance
       - the actual discard IO requests have been moved out of
         transaction commit to a worker thread, improving commit latency
       - IO rate and request size can be tuned by sysfs files, for now
         enabled only with CONFIG_BTRFS_DEBUG as we might need to
         add/delete the files and don't have a stable-ish ABI for
         general use, defaults are conservative

   - export device state info in sysfs, eg. missing, writeable

   - no discard of extents known to be untouched on disk (eg. after
     reservation)

   - device stats reset is logged with process name and PID that called
     the ioctl

  Fixes:

   - fix missing hole after hole punching and fsync when using NO_HOLES

   - writeback: range cyclic mode could miss some dirty pages and lead
     to OOM

   - two more corner cases for metadata_uuid change after power loss
     during the change

   - fix infinite loop during fsync after mix of rename operations

  Core changes:

   - qgroup assign returns ENOTCONN when quotas not enabled, used to
     return EINVAL that was confusing

   - device closing does not need to allocate memory anymore

   - snapshot aware code got removed, disabled for years due to
     performance problems, reimplmentation will allow to select wheter
     defrag breaks or does not break COW on shared extents

   - tree-checker:
       - check leaf chunk item size, cross check against number of
         stripes
       - verify location keys for DIR_ITEM, DIR_INDEX and XATTR items

   - new self test for physical -> logical mapping code, used for super
     block range exclusion

   - assertion helpers/macros updated to avoid objtool "unreachable
     code" reports on older compilers or config option combinations"

* tag 'for-5.6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (84 commits)
  btrfs: free block groups after free'ing fs trees
  btrfs: Fix split-brain handling when changing FSID to metadata uuid
  btrfs: Handle another split brain scenario with metadata uuid feature
  btrfs: Factor out metadata_uuid code from find_fsid.
  btrfs: Call find_fsid from find_fsid_inprogress
  Btrfs: fix infinite loop during fsync after rename operations
  btrfs: set trans->drity in btrfs_commit_transaction
  btrfs: drop log root for dropped roots
  btrfs: sysfs, add devid/dev_state kobject and device attributes
  btrfs: Refactor btrfs_rmap_block to improve readability
  btrfs: Add self-tests for btrfs_rmap_block
  btrfs: selftests: Add support for dummy devices
  btrfs: Move and unexport btrfs_rmap_block
  btrfs: separate definition of assertion failure handlers
  btrfs: device stats, log when stats are zeroed
  btrfs: fix improper setting of scanned for range cyclic write cache pages
  btrfs: safely advance counter when looking up bio csums
  btrfs: remove unused member btrfs_device::work
  btrfs: remove unnecessary wrapper get_alloc_profile
  btrfs: add correction to handle -1 edge case in async discard
  ...

1 files changed, 83 insertions, 751 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index c70baafb2a39..6d2bb58d277a 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -44,7 +44,6 @@
 #include "locking.h"
 #include "free-space-cache.h"
 #include "inode-map.h"
-#include "backref.h"
 #include "props.h"
 #include "qgroup.h"
 #include "delalloc-space.h"
@@ -64,7 +63,6 @@ struct btrfs_dio_data {
 
 static const struct inode_operations btrfs_dir_inode_operations;
 static const struct inode_operations btrfs_symlink_inode_operations;
-static const struct inode_operations btrfs_dir_ro_inode_operations;
 static const struct inode_operations btrfs_special_inode_operations;
 static const struct inode_operations btrfs_file_inode_operations;
 static const struct address_space_operations btrfs_aops;
@@ -2128,7 +2126,7 @@ static blk_status_t btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
 							   bio_flags);
 			goto out;
 		} else if (!skip_sum) {
-			ret = btrfs_lookup_bio_sums(inode, bio, NULL);
+			ret = btrfs_lookup_bio_sums(inode, bio, (u64)-1, NULL);
 			if (ret)
 				goto out;
 		}
@@ -2394,649 +2392,6 @@ out:
 	return ret;
 }
 
-/* snapshot-aware defrag */
-struct sa_defrag_extent_backref {
-	struct rb_node node;
-	struct old_sa_defrag_extent *old;
-	u64 root_id;
-	u64 inum;
-	u64 file_pos;
-	u64 extent_offset;
-	u64 num_bytes;
-	u64 generation;
-};
-
-struct old_sa_defrag_extent {
-	struct list_head list;
-	struct new_sa_defrag_extent *new;
-
-	u64 extent_offset;
-	u64 bytenr;
-	u64 offset;
-	u64 len;
-	int count;
-};
-
-struct new_sa_defrag_extent {
-	struct rb_root root;
-	struct list_head head;
-	struct btrfs_path *path;
-	struct inode *inode;
-	u64 file_pos;
-	u64 len;
-	u64 bytenr;
-	u64 disk_len;
-	u8 compress_type;
-};
-
-static int backref_comp(struct sa_defrag_extent_backref *b1,
-			struct sa_defrag_extent_backref *b2)
-{
-	if (b1->root_id < b2->root_id)
-		return -1;
-	else if (b1->root_id > b2->root_id)
-		return 1;
-
-	if (b1->inum < b2->inum)
-		return -1;
-	else if (b1->inum > b2->inum)
-		return 1;
-
-	if (b1->file_pos < b2->file_pos)
-		return -1;
-	else if (b1->file_pos > b2->file_pos)
-		return 1;
-
-	/*
-	 * [------------------------------] ===> (a range of space)
-	 *     |<--->|   |<---->| =============> (fs/file tree A)
-	 * |<---------------------------->| ===> (fs/file tree B)
-	 *
-	 * A range of space can refer to two file extents in one tree while
-	 * refer to only one file extent in another tree.
-	 *
-	 * So we may process a disk offset more than one time(two extents in A)
-	 * and locate at the same extent(one extent in B), then insert two same
-	 * backrefs(both refer to the extent in B).
-	 */
-	return 0;
-}
-
-static void backref_insert(struct rb_root *root,
-			   struct sa_defrag_extent_backref *backref)
-{
-	struct rb_node **p = &root->rb_node;
-	struct rb_node *parent = NULL;
-	struct sa_defrag_extent_backref *entry;
-	int ret;
-
-	while (*p) {
-		parent = *p;
-		entry = rb_entry(parent, struct sa_defrag_extent_backref, node);
-
-		ret = backref_comp(backref, entry);
-		if (ret < 0)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-
-	rb_link_node(&backref->node, parent, p);
-	rb_insert_color(&backref->node, root);
-}
-
-/*
- * Note the backref might has changed, and in this case we just return 0.
- */
-static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id,
-				       void *ctx)
-{
-	struct btrfs_file_extent_item *extent;
-	struct old_sa_defrag_extent *old = ctx;
-	struct new_sa_defrag_extent *new = old->new;
-	struct btrfs_path *path = new->path;
-	struct btrfs_key key;
-	struct btrfs_root *root;
-	struct sa_defrag_extent_backref *backref;
-	struct extent_buffer *leaf;
-	struct inode *inode = new->inode;
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	int slot;
-	int ret;
-	u64 extent_offset;
-	u64 num_bytes;
-
-	if (BTRFS_I(inode)->root->root_key.objectid == root_id &&
-	    inum == btrfs_ino(BTRFS_I(inode)))
-		return 0;
-
-	key.objectid = root_id;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(root)) {
-		if (PTR_ERR(root) == -ENOENT)
-			return 0;
-		WARN_ON(1);
-		btrfs_debug(fs_info, "inum=%llu, offset=%llu, root_id=%llu",
-			 inum, offset, root_id);
-		return PTR_ERR(root);
-	}
-
-	key.objectid = inum;
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	if (offset > (u64)-1 << 32)
-		key.offset = 0;
-	else
-		key.offset = offset;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (WARN_ON(ret < 0))
-		return ret;
-	ret = 0;
-
-	while (1) {
-		cond_resched();
-
-		leaf = path->nodes[0];
-		slot = path->slots[0];
-
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				goto out;
-			} else if (ret > 0) {
-				ret = 0;
-				goto out;
-			}
-			continue;
-		}
-
-		path->slots[0]++;
-
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-
-		if (key.objectid > inum)
-			goto out;
-
-		if (key.objectid < inum || key.type != BTRFS_EXTENT_DATA_KEY)
-			continue;
-
-		extent = btrfs_item_ptr(leaf, slot,
-					struct btrfs_file_extent_item);
-
-		if (btrfs_file_extent_disk_bytenr(leaf, extent) != old->bytenr)
-			continue;
-
-		/*
-		 * 'offset' refers to the exact key.offset,
-		 * NOT the 'offset' field in btrfs_extent_data_ref, ie.
-		 * (key.offset - extent_offset).
-		 */
-		if (key.offset != offset)
-			continue;
-
-		extent_offset = btrfs_file_extent_offset(leaf, extent);
-		num_bytes = btrfs_file_extent_num_bytes(leaf, extent);
-
-		if (extent_offset >= old->extent_offset + old->offset +
-		    old->len || extent_offset + num_bytes <=
-		    old->extent_offset + old->offset)
-			continue;
-		break;
-	}
-
-	backref = kmalloc(sizeof(*backref), GFP_NOFS);
-	if (!backref) {
-		ret = -ENOENT;
-		goto out;
-	}
-
-	backref->root_id = root_id;
-	backref->inum = inum;
-	backref->file_pos = offset;
-	backref->num_bytes = num_bytes;
-	backref->extent_offset = extent_offset;
-	backref->generation = btrfs_file_extent_generation(leaf, extent);
-	backref->old = old;
-	backref_insert(&new->root, backref);
-	old->count++;
-out:
-	btrfs_release_path(path);
-	WARN_ON(ret);
-	return ret;
-}
-
-static noinline bool record_extent_backrefs(struct btrfs_path *path,
-				   struct new_sa_defrag_extent *new)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb);
-	struct old_sa_defrag_extent *old, *tmp;
-	int ret;
-
-	new->path = path;
-
-	list_for_each_entry_safe(old, tmp, &new->head, list) {
-		ret = iterate_inodes_from_logical(old->bytenr +
-						  old->extent_offset, fs_info,
-						  path, record_one_backref,
-						  old, false);
-		if (ret < 0 && ret != -ENOENT)
-			return false;
-
-		/* no backref to be processed for this extent */
-		if (!old->count) {
-			list_del(&old->list);
-			kfree(old);
-		}
-	}
-
-	if (list_empty(&new->head))
-		return false;
-
-	return true;
-}
-
-static int relink_is_mergable(struct extent_buffer *leaf,
-			      struct btrfs_file_extent_item *fi,
-			      struct new_sa_defrag_extent *new)
-{
-	if (btrfs_file_extent_disk_bytenr(leaf, fi) != new->bytenr)
-		return 0;
-
-	if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG)
-		return 0;
-
-	if (btrfs_file_extent_compression(leaf, fi) != new->compress_type)
-		return 0;
-
-	if (btrfs_file_extent_encryption(leaf, fi) ||
-	    btrfs_file_extent_other_encoding(leaf, fi))
-		return 0;
-
-	return 1;
-}
-
-/*
- * Note the backref might has changed, and in this case we just return 0.
- */
-static noinline int relink_extent_backref(struct btrfs_path *path,
-				 struct sa_defrag_extent_backref *prev,
-				 struct sa_defrag_extent_backref *backref)
-{
-	struct btrfs_file_extent_item *extent;
-	struct btrfs_file_extent_item *item;
-	struct btrfs_ordered_extent *ordered;
-	struct btrfs_trans_handle *trans;
-	struct btrfs_ref ref = { 0 };
-	struct btrfs_root *root;
-	struct btrfs_key key;
-	struct extent_buffer *leaf;
-	struct old_sa_defrag_extent *old = backref->old;
-	struct new_sa_defrag_extent *new = old->new;
-	struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb);
-	struct inode *inode;
-	struct extent_state *cached = NULL;
-	int ret = 0;
-	u64 start;
-	u64 len;
-	u64 lock_start;
-	u64 lock_end;
-	bool merge = false;
-	int index;
-
-	if (prev && prev->root_id == backref->root_id &&
-	    prev->inum == backref->inum &&
-	    prev->file_pos + prev->num_bytes == backref->file_pos)
-		merge = true;
-
-	/* step 1: get root */
-	key.objectid = backref->root_id;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
-		if (PTR_ERR(root) == -ENOENT)
-			return 0;
-		return PTR_ERR(root);
-	}
-
-	if (btrfs_root_readonly(root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
-		return 0;
-	}
-
-	/* step 2: get inode */
-	key.objectid = backref->inum;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	inode = btrfs_iget(fs_info->sb, &key, root);
-	if (IS_ERR(inode)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
-		return 0;
-	}
-
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
-
-	/* step 3: relink backref */
-	lock_start = backref->file_pos;
-	lock_end = backref->file_pos + backref->num_bytes - 1;
-	lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
-			 &cached);
-
-	ordered = btrfs_lookup_first_ordered_extent(inode, lock_end);
-	if (ordered) {
-		btrfs_put_ordered_extent(ordered);
-		goto out_unlock;
-	}
-
-	trans = btrfs_join_transaction(root);
-	if (IS_ERR(trans)) {
-		ret = PTR_ERR(trans);
-		goto out_unlock;
-	}
-
-	key.objectid = backref->inum;
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = backref->file_pos;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0) {
-		goto out_free_path;
-	} else if (ret > 0) {
-		ret = 0;
-		goto out_free_path;
-	}
-
-	extent = btrfs_item_ptr(path->nodes[0], path->slots[0],
-				struct btrfs_file_extent_item);
-
-	if (btrfs_file_extent_generation(path->nodes[0], extent) !=
-	    backref->generation)
-		goto out_free_path;
-
-	btrfs_release_path(path);
-
-	start = backref->file_pos;
-	if (backref->extent_offset < old->extent_offset + old->offset)
-		start += old->extent_offset + old->offset -
-			 backref->extent_offset;
-
-	len = min(backref->extent_offset + backref->num_bytes,
-		  old->extent_offset + old->offset + old->len);
-	len -= max(backref->extent_offset, old->extent_offset + old->offset);
-
-	ret = btrfs_drop_extents(trans, root, inode, start,
-				 start + len, 1);
-	if (ret)
-		goto out_free_path;
-again:
-	key.objectid = btrfs_ino(BTRFS_I(inode));
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = start;
-
-	path->leave_spinning = 1;
-	if (merge) {
-		struct btrfs_file_extent_item *fi;
-		u64 extent_len;
-		struct btrfs_key found_key;
-
-		ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
-		if (ret < 0)
-			goto out_free_path;
-
-		path->slots[0]--;
-		leaf = path->nodes[0];
-		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
-
-		fi = btrfs_item_ptr(leaf, path->slots[0],
-				    struct btrfs_file_extent_item);
-		extent_len = btrfs_file_extent_num_bytes(leaf, fi);
-
-		if (extent_len + found_key.offset == start &&
-		    relink_is_mergable(leaf, fi, new)) {
-			btrfs_set_file_extent_num_bytes(leaf, fi,
-							extent_len + len);
-			btrfs_mark_buffer_dirty(leaf);
-			inode_add_bytes(inode, len);
-
-			ret = 1;
-			goto out_free_path;
-		} else {
-			merge = false;
-			btrfs_release_path(path);
-			goto again;
-		}
-	}
-
-	ret = btrfs_insert_empty_item(trans, root, path, &key,
-					sizeof(*extent));
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		goto out_free_path;
-	}
-
-	leaf = path->nodes[0];
-	item = btrfs_item_ptr(leaf, path->slots[0],
-				struct btrfs_file_extent_item);
-	btrfs_set_file_extent_disk_bytenr(leaf, item, new->bytenr);
-	btrfs_set_file_extent_disk_num_bytes(leaf, item, new->disk_len);
-	btrfs_set_file_extent_offset(leaf, item, start - new->file_pos);
-	btrfs_set_file_extent_num_bytes(leaf, item, len);
-	btrfs_set_file_extent_ram_bytes(leaf, item, new->len);
-	btrfs_set_file_extent_generation(leaf, item, trans->transid);
-	btrfs_set_file_extent_type(leaf, item, BTRFS_FILE_EXTENT_REG);
-	btrfs_set_file_extent_compression(leaf, item, new->compress_type);
-	btrfs_set_file_extent_encryption(leaf, item, 0);
-	btrfs_set_file_extent_other_encoding(leaf, item, 0);
-
-	btrfs_mark_buffer_dirty(leaf);
-	inode_add_bytes(inode, len);
-	btrfs_release_path(path);
-
-	btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, new->bytenr,
-			       new->disk_len, 0);
-	btrfs_init_data_ref(&ref, backref->root_id, backref->inum,
-			    new->file_pos);  /* start - extent_offset */
-	ret = btrfs_inc_extent_ref(trans, &ref);
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		goto out_free_path;
-	}
-
-	ret = 1;
-out_free_path:
-	btrfs_release_path(path);
-	path->leave_spinning = 0;
-	btrfs_end_transaction(trans);
-out_unlock:
-	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
-			     &cached);
-	iput(inode);
-	return ret;
-}
-
-static void free_sa_defrag_extent(struct new_sa_defrag_extent *new)
-{
-	struct old_sa_defrag_extent *old, *tmp;
-
-	if (!new)
-		return;
-
-	list_for_each_entry_safe(old, tmp, &new->head, list) {
-		kfree(old);
-	}
-	kfree(new);
-}
-
-static void relink_file_extents(struct new_sa_defrag_extent *new)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb);
-	struct btrfs_path *path;
-	struct sa_defrag_extent_backref *backref;
-	struct sa_defrag_extent_backref *prev = NULL;
-	struct rb_node *node;
-	int ret;
-
-	path = btrfs_alloc_path();
-	if (!path)
-		return;
-
-	if (!record_extent_backrefs(path, new)) {
-		btrfs_free_path(path);
-		goto out;
-	}
-	btrfs_release_path(path);
-
-	while (1) {
-		node = rb_first(&new->root);
-		if (!node)
-			break;
-		rb_erase(node, &new->root);
-
-		backref = rb_entry(node, struct sa_defrag_extent_backref, node);
-
-		ret = relink_extent_backref(path, prev, backref);
-		WARN_ON(ret < 0);
-
-		kfree(prev);
-
-		if (ret == 1)
-			prev = backref;
-		else
-			prev = NULL;
-		cond_resched();
-	}
-	kfree(prev);
-
-	btrfs_free_path(path);
-out:
-	free_sa_defrag_extent(new);
-
-	atomic_dec(&fs_info->defrag_running);
-	wake_up(&fs_info->transaction_wait);
-}
-
-static struct new_sa_defrag_extent *
-record_old_file_extents(struct inode *inode,
-			struct btrfs_ordered_extent *ordered)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_path *path;
-	struct btrfs_key key;
-	struct old_sa_defrag_extent *old;
-	struct new_sa_defrag_extent *new;
-	int ret;
-
-	new = kmalloc(sizeof(*new), GFP_NOFS);
-	if (!new)
-		return NULL;
-
-	new->inode = inode;
-	new->file_pos = ordered->file_offset;
-	new->len = ordered->len;
-	new->bytenr = ordered->start;
-	new->disk_len = ordered->disk_len;
-	new->compress_type = ordered->compress_type;
-	new->root = RB_ROOT;
-	INIT_LIST_HEAD(&new->head);
-
-	path = btrfs_alloc_path();
-	if (!path)
-		goto out_kfree;
-
-	key.objectid = btrfs_ino(BTRFS_I(inode));
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = new->file_pos;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out_free_path;
-	if (ret > 0 && path->slots[0] > 0)
-		path->slots[0]--;
-
-	/* find out all the old extents for the file range */
-	while (1) {
-		struct btrfs_file_extent_item *extent;
-		struct extent_buffer *l;
-		int slot;
-		u64 num_bytes;
-		u64 offset;
-		u64 end;
-		u64 disk_bytenr;
-		u64 extent_offset;
-
-		l = path->nodes[0];
-		slot = path->slots[0];
-
-		if (slot >= btrfs_header_nritems(l)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto out_free_path;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(l, &key, slot);
-
-		if (key.objectid != btrfs_ino(BTRFS_I(inode)))
-			break;
-		if (key.type != BTRFS_EXTENT_DATA_KEY)
-			break;
-		if (key.offset >= new->file_pos + new->len)
-			break;
-
-		extent = btrfs_item_ptr(l, slot, struct btrfs_file_extent_item);
-
-		num_bytes = btrfs_file_extent_num_bytes(l, extent);
-		if (key.offset + num_bytes < new->file_pos)
-			goto next;
-
-		disk_bytenr = btrfs_file_extent_disk_bytenr(l, extent);
-		if (!disk_bytenr)
-			goto next;
-
-		extent_offset = btrfs_file_extent_offset(l, extent);
-
-		old = kmalloc(sizeof(*old), GFP_NOFS);
-		if (!old)
-			goto out_free_path;
-
-		offset = max(new->file_pos, key.offset);
-		end = min(new->file_pos + new->len, key.offset + num_bytes);
-
-		old->bytenr = disk_bytenr;
-		old->extent_offset = extent_offset;
-		old->offset = offset - key.offset;
-		old->len = end - offset;
-		old->new = new;
-		old->count = 0;
-		list_add_tail(&old->list, &new->head);
-next:
-		path->slots[0]++;
-		cond_resched();
-	}
-
-	btrfs_free_path(path);
-	atomic_inc(&fs_info->defrag_running);
-
-	return new;
-
-out_free_path:
-	btrfs_free_path(path);
-out_kfree:
-	free_sa_defrag_extent(new);
-	return NULL;
-}
-
 static void btrfs_release_delalloc_bytes(struct btrfs_fs_info *fs_info,
 					 u64 start, u64 len)
 {
@@ -3064,15 +2419,19 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	struct btrfs_trans_handle *trans = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_state *cached_state = NULL;
-	struct new_sa_defrag_extent *new = NULL;
+	u64 start, end;
 	int compress_type = 0;
 	int ret = 0;
-	u64 logical_len = ordered_extent->len;
+	u64 logical_len = ordered_extent->num_bytes;
 	bool freespace_inode;
 	bool truncated = false;
 	bool range_locked = false;
 	bool clear_new_delalloc_bytes = false;
 	bool clear_reserved_extent = true;
+	unsigned int clear_bits;
+
+	start = ordered_extent->file_offset;
+	end = start + ordered_extent->num_bytes - 1;
 
 	if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
 	    !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags) &&
@@ -3086,10 +2445,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		goto out;
 	}
 
-	btrfs_free_io_failure_record(BTRFS_I(inode),
-			ordered_extent->file_offset,
-			ordered_extent->file_offset +
-			ordered_extent->len - 1);
+	btrfs_free_io_failure_record(BTRFS_I(inode), start, end);
 
 	if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) {
 		truncated = true;
@@ -3107,8 +2463,8 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		 * space for NOCOW range.
 		 * As NOCOW won't cause a new delayed ref, just free the space
 		 */
-		btrfs_qgroup_free_data(inode, NULL, ordered_extent->file_offset,
-				       ordered_extent->len);
+		btrfs_qgroup_free_data(inode, NULL, start,
+				       ordered_extent->num_bytes);
 		btrfs_ordered_update_i_size(inode, 0, ordered_extent);
 		if (freespace_inode)
 			trans = btrfs_join_transaction_spacecache(root);
@@ -3127,23 +2483,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	}
 
 	range_locked = true;
-	lock_extent_bits(io_tree, ordered_extent->file_offset,
-			 ordered_extent->file_offset + ordered_extent->len - 1,
-			 &cached_state);
-
-	ret = test_range_bit(io_tree, ordered_extent->file_offset,
-			ordered_extent->file_offset + ordered_extent->len - 1,
-			EXTENT_DEFRAG, 0, cached_state);
-	if (ret) {
-		u64 last_snapshot = btrfs_root_last_snapshot(&root->root_item);
-		if (0 && last_snapshot >= BTRFS_I(inode)->generation)
-			/* the inode is shared */
-			new = record_old_file_extents(inode, ordered_extent);
-
-		clear_extent_bit(io_tree, ordered_extent->file_offset,
-			ordered_extent->file_offset + ordered_extent->len - 1,
-			EXTENT_DEFRAG, 0, 0, &cached_state);
-	}
+	lock_extent_bits(io_tree, start, end, &cached_state);
 
 	if (freespace_inode)
 		trans = btrfs_join_transaction_spacecache(root);
@@ -3161,31 +2501,30 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		compress_type = ordered_extent->compress_type;
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
 		BUG_ON(compress_type);
-		btrfs_qgroup_free_data(inode, NULL, ordered_extent->file_offset,
-				       ordered_extent->len);
+		btrfs_qgroup_free_data(inode, NULL, start,
+				       ordered_extent->num_bytes);
 		ret = btrfs_mark_extent_written(trans, BTRFS_I(inode),
 						ordered_extent->file_offset,
 						ordered_extent->file_offset +
 						logical_len);
 	} else {
 		BUG_ON(root == fs_info->tree_root);
-		ret = insert_reserved_file_extent(trans, inode,
-						ordered_extent->file_offset,
-						ordered_extent->start,
-						ordered_extent->disk_len,
+		ret = insert_reserved_file_extent(trans, inode, start,
+						ordered_extent->disk_bytenr,
+						ordered_extent->disk_num_bytes,
 						logical_len, logical_len,
 						compress_type, 0, 0,
 						BTRFS_FILE_EXTENT_REG);
 		if (!ret) {
 			clear_reserved_extent = false;
 			btrfs_release_delalloc_bytes(fs_info,
-						     ordered_extent->start,
-						     ordered_extent->disk_len);
+						ordered_extent->disk_bytenr,
+						ordered_extent->disk_num_bytes);
 		}
 	}
 	unpin_extent_cache(&BTRFS_I(inode)->extent_tree,
-			   ordered_extent->file_offset, ordered_extent->len,
-			   trans->transid);
+			   ordered_extent->file_offset,
+			   ordered_extent->num_bytes, trans->transid);
 	if (ret < 0) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
@@ -3205,37 +2544,27 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	}
 	ret = 0;
 out:
-	if (range_locked || clear_new_delalloc_bytes) {
-		unsigned int clear_bits = 0;
-
-		if (range_locked)
-			clear_bits |= EXTENT_LOCKED;
-		if (clear_new_delalloc_bytes)
-			clear_bits |= EXTENT_DELALLOC_NEW;
-		clear_extent_bit(&BTRFS_I(inode)->io_tree,
-				 ordered_extent->file_offset,
-				 ordered_extent->file_offset +
-				 ordered_extent->len - 1,
-				 clear_bits,
-				 (clear_bits & EXTENT_LOCKED) ? 1 : 0,
-				 0, &cached_state);
-	}
+	clear_bits = EXTENT_DEFRAG;
+	if (range_locked)
+		clear_bits |= EXTENT_LOCKED;
+	if (clear_new_delalloc_bytes)
+		clear_bits |= EXTENT_DELALLOC_NEW;
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits,
+			 (clear_bits & EXTENT_LOCKED) ? 1 : 0, 0,
+			 &cached_state);
 
 	if (trans)
 		btrfs_end_transaction(trans);
 
 	if (ret || truncated) {
-		u64 start, end;
+		u64 unwritten_start = start;
 
 		if (truncated)
-			start = ordered_extent->file_offset + logical_len;
-		else
-			start = ordered_extent->file_offset;
-		end = ordered_extent->file_offset + ordered_extent->len - 1;
-		clear_extent_uptodate(io_tree, start, end, NULL);
+			unwritten_start += logical_len;
+		clear_extent_uptodate(io_tree, unwritten_start, end, NULL);
 
 		/* Drop the cache for the part of the extent we didn't write. */
-		btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 0);
+		btrfs_drop_extent_cache(BTRFS_I(inode), unwritten_start, end, 0);
 
 		/*
 		 * If the ordered extent had an IOERR or something else went
@@ -3250,29 +2579,28 @@ out:
 		if ((ret || !logical_len) &&
 		    clear_reserved_extent &&
 		    !test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
-		    !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags))
+		    !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
+			/*
+			 * Discard the range before returning it back to the
+			 * free space pool
+			 */
+			if (ret && btrfs_test_opt(fs_info, DISCARD_SYNC))
+				btrfs_discard_extent(fs_info,
+						ordered_extent->disk_bytenr,
+						ordered_extent->disk_num_bytes,
+						NULL);
 			btrfs_free_reserved_extent(fs_info,
-						   ordered_extent->start,
-						   ordered_extent->disk_len, 1);
+					ordered_extent->disk_bytenr,
+					ordered_extent->disk_num_bytes, 1);
+		}
 	}
 
-
 	/*
 	 * This needs to be done to make sure anybody waiting knows we are done
 	 * updating everything for this ordered extent.
 	 */
 	btrfs_remove_ordered_extent(inode, ordered_extent);
 
-	/* for snapshot-aware defrag */
-	if (new) {
-		if (ret) {
-			free_sa_defrag_extent(new);
-			atomic_dec(&fs_info->defrag_running);
-		} else {
-			relink_file_extents(new);
-		}
-	}
-
 	/* once for us */
 	btrfs_put_ordered_extent(ordered_extent);
 	/* once for the tree */
@@ -5176,7 +4504,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 	cur_offset = hole_start;
 	while (1) {
 		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset,
-				block_end - cur_offset, 0);
+				      block_end - cur_offset);
 		if (IS_ERR(em)) {
 			err = PTR_ERR(em);
 			em = NULL;
@@ -5860,7 +5188,11 @@ static struct inode *new_simple_dir(struct super_block *s,
 	set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags);
 
 	inode->i_ino = BTRFS_EMPTY_SUBVOL_DIR_OBJECTID;
-	inode->i_op = &btrfs_dir_ro_inode_operations;
+	/*
+	 * We only need lookup, the rest is read-only and there's no inode
+	 * associated with the dentry
+	 */
+	inode->i_op = &simple_dir_inode_operations;
 	inode->i_opflags &= ~IOP_XATTR;
 	inode->i_fop = &simple_dir_operations;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO;
@@ -6951,18 +6283,27 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	return ret;
 }
 
-/*
- * a bit scary, this does extent mapping from logical file offset to the disk.
- * the ugly parts come from merging extents from the disk with the in-ram
- * representation.  This gets more complex because of the data=ordered code,
- * where the in-ram extents might be locked pending data=ordered completion.
+/**
+ * btrfs_get_extent - Lookup the first extent overlapping a range in a file.
+ * @inode:	file to search in
+ * @page:	page to read extent data into if the extent is inline
+ * @pg_offset:	offset into @page to copy to
+ * @start:	file offset
+ * @len:	length of range starting at @start
+ *
+ * This returns the first &struct extent_map which overlaps with the given
+ * range, reading it from the B-tree and caching it if necessary. Note that
+ * there may be more extents which overlap the given range after the returned
+ * extent_map.
  *
- * This also copies inline extents directly into the page.
+ * If @page is not NULL and the extent is inline, this also reads the extent
+ * data directly into the page and marks the extent up to date in the io_tree.
+ *
+ * Return: ERR_PTR on error, non-NULL extent_map on success.
  */
 struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
-				    struct page *page,
-				    size_t pg_offset, u64 start, u64 len,
-				    int create)
+				    struct page *page, size_t pg_offset,
+				    u64 start, u64 len)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	int ret;
@@ -6979,7 +6320,6 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	struct extent_map *em = NULL;
 	struct extent_map_tree *em_tree = &inode->extent_tree;
 	struct extent_io_tree *io_tree = &inode->io_tree;
-	const bool new_inline = !page || create;
 
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree, start, len);
@@ -7102,8 +6442,7 @@ next:
 		goto insert;
 	}
 
-	btrfs_extent_item_to_extent_map(inode, path, item,
-			new_inline, em);
+	btrfs_extent_item_to_extent_map(inode, path, item, !page, em);
 
 	if (extent_type == BTRFS_FILE_EXTENT_REG ||
 	    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
@@ -7115,7 +6454,7 @@ next:
 		size_t extent_offset;
 		size_t copy_size;
 
-		if (new_inline)
+		if (!page)
 			goto out;
 
 		size = btrfs_file_extent_ram_bytes(leaf, item);
@@ -7198,7 +6537,7 @@ struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
 	u64 delalloc_end;
 	int err = 0;
 
-	em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
+	em = btrfs_get_extent(inode, NULL, 0, start, len);
 	if (IS_ERR(em))
 		return em;
 	/*
@@ -7823,7 +7162,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
 		goto err;
 	}
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len);
 	if (IS_ERR(em)) {
 		ret = PTR_ERR(em);
 		goto unlock_err;
@@ -8375,8 +7714,8 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode,
 	 * contention.
 	 */
 	if (dip->logical_offset == file_offset) {
-		ret = btrfs_lookup_bio_sums_dio(inode, dip->orig_bio,
-						file_offset);
+		ret = btrfs_lookup_bio_sums(inode, dip->orig_bio, file_offset,
+					    NULL);
 		if (ret)
 			return ret;
 	}
@@ -8889,7 +8228,8 @@ again:
 	ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
 					page_end - start + 1);
 	if (ordered) {
-		end = min(page_end, ordered->file_offset + ordered->len - 1);
+		end = min(page_end,
+			  ordered->file_offset + ordered->num_bytes - 1);
 		/*
 		 * IO on this page will never be started, so we need
 		 * to account for any ordered extents now
@@ -9090,7 +8430,6 @@ again:
 		ret = VM_FAULT_SIGBUS;
 		goto out_unlock;
 	}
-	ret2 = 0;
 
 	/* page is wholly or partially inside EOF */
 	if (page_start + PAGE_SIZE > size)
@@ -9114,12 +8453,10 @@ again:
 
 	unlock_extent_cached(io_tree, page_start, page_end, &cached_state);
 
-	if (!ret2) {
-		btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
-		sb_end_pagefault(inode->i_sb);
-		extent_changeset_free(data_reserved);
-		return VM_FAULT_LOCKED;
-	}
+	btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
+	sb_end_pagefault(inode->i_sb);
+	extent_changeset_free(data_reserved);
+	return VM_FAULT_LOCKED;
 
 out_unlock:
 	unlock_page(page);
@@ -9417,7 +8754,7 @@ void btrfs_destroy_inode(struct inode *inode)
 		else {
 			btrfs_err(fs_info,
 				  "found ordered extent %llu %llu on inode cleanup",
-				  ordered->file_offset, ordered->len);
+				  ordered->file_offset, ordered->num_bytes);
 			btrfs_remove_ordered_extent(inode, ordered);
 			btrfs_put_ordered_extent(ordered);
 			btrfs_put_ordered_extent(ordered);
@@ -10836,7 +10173,7 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 		struct btrfs_block_group *bg;
 		u64 len = isize - start;
 
-		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
+		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
 			goto out;
@@ -11004,11 +10341,6 @@ static const struct inode_operations btrfs_dir_inode_operations = {
 	.update_time	= btrfs_update_time,
 	.tmpfile        = btrfs_tmpfile,
 };
-static const struct inode_operations btrfs_dir_ro_inode_operations = {
-	.lookup		= btrfs_lookup,
-	.permission	= btrfs_permission,
-	.update_time	= btrfs_update_time,
-};
 
 static const struct file_operations btrfs_dir_file_operations = {
 	.llseek		= generic_file_llseek,