1 files changed, 400 insertions, 394 deletions

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 87355a38a654..0e41459b8de6 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -452,46 +452,6 @@ static void btrfs_drop_pages(struct page **pages, size_t num_pages)
 	}
 }
 
-static int btrfs_find_new_delalloc_bytes(struct btrfs_inode *inode,
-					 const u64 start,
-					 const u64 len,
-					 struct extent_state **cached_state)
-{
-	u64 search_start = start;
-	const u64 end = start + len - 1;
-
-	while (search_start < end) {
-		const u64 search_len = end - search_start + 1;
-		struct extent_map *em;
-		u64 em_len;
-		int ret = 0;
-
-		em = btrfs_get_extent(inode, NULL, 0, search_start, search_len);
-		if (IS_ERR(em))
-			return PTR_ERR(em);
-
-		if (em->block_start != EXTENT_MAP_HOLE)
-			goto next;
-
-		em_len = em->len;
-		if (em->start < search_start)
-			em_len -= search_start - em->start;
-		if (em_len > search_len)
-			em_len = search_len;
-
-		ret = set_extent_bit(&inode->io_tree, search_start,
-				     search_start + em_len - 1,
-				     EXTENT_DELALLOC_NEW,
-				     NULL, cached_state, GFP_NOFS);
-next:
-		search_start = extent_map_end(em);
-		free_extent_map(em);
-		if (ret)
-			return ret;
-	}
-	return 0;
-}
-
 /*
  * after copy_from_user, pages need to be dirtied and we need to make
  * sure holes are created between the current EOF and the start of
@@ -502,7 +462,7 @@ next:
  */
 int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages,
 		      size_t num_pages, loff_t pos, size_t write_bytes,
-		      struct extent_state **cached)
+		      struct extent_state **cached, bool noreserve)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	int err = 0;
@@ -514,7 +474,13 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages,
 	loff_t isize = i_size_read(&inode->vfs_inode);
 	unsigned int extra_bits = 0;
 
-	start_pos = pos & ~((u64) fs_info->sectorsize - 1);
+	if (write_bytes == 0)
+		return 0;
+
+	if (noreserve)
+		extra_bits |= EXTENT_NORESERVE;
+
+	start_pos = round_down(pos, fs_info->sectorsize);
 	num_bytes = round_up(write_bytes + pos - start_pos,
 			     fs_info->sectorsize);
 
@@ -528,23 +494,6 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages,
 			 EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
 			 0, 0, cached);
 
-	if (!btrfs_is_free_space_inode(inode)) {
-		if (start_pos >= isize &&
-		    !(inode->flags & BTRFS_INODE_PREALLOC)) {
-			/*
-			 * There can't be any extents following eof in this case
-			 * so just set the delalloc new bit for the range
-			 * directly.
-			 */
-			extra_bits |= EXTENT_DELALLOC_NEW;
-		} else {
-			err = btrfs_find_new_delalloc_bytes(inode, start_pos,
-							    num_bytes, cached);
-			if (err)
-				return err;
-		}
-	}
-
 	err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block,
 					extra_bits, cached);
 	if (err)
@@ -728,14 +677,16 @@ next:
  * If an extent intersects the range but is not entirely inside the range
  * it is either truncated or split.  Anything entirely inside the range
  * is deleted from the tree.
+ *
+ * Note: the VFS' inode number of bytes is not updated, it's up to the caller
+ * to deal with that. We set the field 'bytes_found' of the arguments structure
+ * with the number of allocated bytes found in the target range, so that the
+ * caller can update the inode's number of bytes in an atomic way when
+ * replacing extents in a range to avoid races with stat(2).
  */
-int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
-			 struct btrfs_root *root, struct btrfs_inode *inode,
-			 struct btrfs_path *path, u64 start, u64 end,
-			 u64 *drop_end, int drop_cache,
-			 int replace_extent,
-			 u32 extent_item_size,
-			 int *key_inserted)
+int btrfs_drop_extents(struct btrfs_trans_handle *trans,
+		       struct btrfs_root *root, struct btrfs_inode *inode,
+		       struct btrfs_drop_extents_args *args)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *leaf;
@@ -743,14 +694,13 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
 	struct btrfs_ref ref = { 0 };
 	struct btrfs_key key;
 	struct btrfs_key new_key;
-	struct inode *vfs_inode = &inode->vfs_inode;
 	u64 ino = btrfs_ino(inode);
-	u64 search_start = start;
+	u64 search_start = args->start;
 	u64 disk_bytenr = 0;
 	u64 num_bytes = 0;
 	u64 extent_offset = 0;
 	u64 extent_end = 0;
-	u64 last_end = start;
+	u64 last_end = args->start;
 	int del_nr = 0;
 	int del_slot = 0;
 	int extent_type;
@@ -760,11 +710,26 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
 	int update_refs;
 	int found = 0;
 	int leafs_visited = 0;
+	struct btrfs_path *path = args->path;
+
+	args->bytes_found = 0;
+	args->extent_inserted = false;
+
+	/* Must always have a path if ->replace_extent is true */
+	ASSERT(!(args->replace_extent && !args->path));
+
+	if (!path) {
+		path = btrfs_alloc_path();
+		if (!path) {
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
 
-	if (drop_cache)
-		btrfs_drop_extent_cache(inode, start, end - 1, 0);
+	if (args->drop_cache)
+		btrfs_drop_extent_cache(inode, args->start, args->end - 1, 0);
 
-	if (start >= inode->disk_i_size && !replace_extent)
+	if (args->start >= inode->disk_i_size && !args->replace_extent)
 		modify_tree = 0;
 
 	update_refs = (test_bit(BTRFS_ROOT_SHAREABLE, &root->state) ||
@@ -775,7 +740,7 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
 					       search_start, modify_tree);
 		if (ret < 0)
 			break;
-		if (ret > 0 && path->slots[0] > 0 && search_start == start) {
+		if (ret > 0 && path->slots[0] > 0 && search_start == args->start) {
 			leaf = path->nodes[0];
 			btrfs_item_key_to_cpu(leaf, &key, path->slots[0] - 1);
 			if (key.objectid == ino &&
@@ -810,7 +775,7 @@ next_slot:
 			path->slots[0]++;
 			goto next_slot;
 		}
-		if (key.type > BTRFS_EXTENT_DATA_KEY || key.offset >= end)
+		if (key.type > BTRFS_EXTENT_DATA_KEY || key.offset >= args->end)
 			break;
 
 		fi = btrfs_item_ptr(leaf, path->slots[0],
@@ -852,7 +817,7 @@ next_slot:
 		}
 
 		found = 1;
-		search_start = max(key.offset, start);
+		search_start = max(key.offset, args->start);
 		if (recow || !modify_tree) {
 			modify_tree = -1;
 			btrfs_release_path(path);
@@ -863,7 +828,7 @@ next_slot:
 		 *     | - range to drop - |
 		 *  | -------- extent -------- |
 		 */
-		if (start > key.offset && end < extent_end) {
+		if (args->start > key.offset && args->end < extent_end) {
 			BUG_ON(del_nr > 0);
 			if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 				ret = -EOPNOTSUPP;
@@ -871,7 +836,7 @@ next_slot:
 			}
 
 			memcpy(&new_key, &key, sizeof(new_key));
-			new_key.offset = start;
+			new_key.offset = args->start;
 			ret = btrfs_duplicate_item(trans, root, path,
 						   &new_key);
 			if (ret == -EAGAIN) {
@@ -885,15 +850,15 @@ next_slot:
 			fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
 					    struct btrfs_file_extent_item);
 			btrfs_set_file_extent_num_bytes(leaf, fi,
-							start - key.offset);
+							args->start - key.offset);
 
 			fi = btrfs_item_ptr(leaf, path->slots[0],
 					    struct btrfs_file_extent_item);
 
-			extent_offset += start - key.offset;
+			extent_offset += args->start - key.offset;
 			btrfs_set_file_extent_offset(leaf, fi, extent_offset);
 			btrfs_set_file_extent_num_bytes(leaf, fi,
-							extent_end - start);
+							extent_end - args->start);
 			btrfs_mark_buffer_dirty(leaf);
 
 			if (update_refs && disk_bytenr > 0) {
@@ -903,11 +868,11 @@ next_slot:
 				btrfs_init_data_ref(&ref,
 						root->root_key.objectid,
 						new_key.objectid,
-						start - extent_offset);
+						args->start - extent_offset);
 				ret = btrfs_inc_extent_ref(trans, &ref);
 				BUG_ON(ret); /* -ENOMEM */
 			}
-			key.offset = start;
+			key.offset = args->start;
 		}
 		/*
 		 * From here on out we will have actually dropped something, so
@@ -919,23 +884,23 @@ next_slot:
 		 *  | ---- range to drop ----- |
 		 *      | -------- extent -------- |
 		 */
-		if (start <= key.offset && end < extent_end) {
+		if (args->start <= key.offset && args->end < extent_end) {
 			if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 				ret = -EOPNOTSUPP;
 				break;
 			}
 
 			memcpy(&new_key, &key, sizeof(new_key));
-			new_key.offset = end;
+			new_key.offset = args->end;
 			btrfs_set_item_key_safe(fs_info, path, &new_key);
 
-			extent_offset += end - key.offset;
+			extent_offset += args->end - key.offset;
 			btrfs_set_file_extent_offset(leaf, fi, extent_offset);
 			btrfs_set_file_extent_num_bytes(leaf, fi,
-							extent_end - end);
+							extent_end - args->end);
 			btrfs_mark_buffer_dirty(leaf);
 			if (update_refs && disk_bytenr > 0)
-				inode_sub_bytes(vfs_inode, end - key.offset);
+				args->bytes_found += args->end - key.offset;
 			break;
 		}
 
@@ -944,7 +909,7 @@ next_slot:
 		 *       | ---- range to drop ----- |
 		 *  | -------- extent -------- |
 		 */
-		if (start > key.offset && end >= extent_end) {
+		if (args->start > key.offset && args->end >= extent_end) {
 			BUG_ON(del_nr > 0);
 			if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 				ret = -EOPNOTSUPP;
@@ -952,11 +917,11 @@ next_slot:
 			}
 
 			btrfs_set_file_extent_num_bytes(leaf, fi,
-							start - key.offset);
+							args->start - key.offset);
 			btrfs_mark_buffer_dirty(leaf);
 			if (update_refs && disk_bytenr > 0)
-				inode_sub_bytes(vfs_inode, extent_end - start);
-			if (end == extent_end)
+				args->bytes_found += extent_end - args->start;
+			if (args->end == extent_end)
 				break;
 
 			path->slots[0]++;
@@ -967,7 +932,7 @@ next_slot:
 		 *  | ---- range to drop ----- |
 		 *    | ------ extent ------ |
 		 */
-		if (start <= key.offset && end >= extent_end) {
+		if (args->start <= key.offset && args->end >= extent_end) {
 delete_extent_item:
 			if (del_nr == 0) {
 				del_slot = path->slots[0];
@@ -979,8 +944,7 @@ delete_extent_item:
 
 			if (update_refs &&
 			    extent_type == BTRFS_FILE_EXTENT_INLINE) {
-				inode_sub_bytes(vfs_inode,
-						extent_end - key.offset);
+				args->bytes_found += extent_end - key.offset;
 				extent_end = ALIGN(extent_end,
 						   fs_info->sectorsize);
 			} else if (update_refs && disk_bytenr > 0) {
@@ -993,11 +957,10 @@ delete_extent_item:
 						key.offset - extent_offset);
 				ret = btrfs_free_extent(trans, &ref);
 				BUG_ON(ret); /* -ENOMEM */
-				inode_sub_bytes(vfs_inode,
-						extent_end - key.offset);
+				args->bytes_found += extent_end - key.offset;
 			}
 
-			if (end == extent_end)
+			if (args->end == extent_end)
 				break;
 
 			if (path->slots[0] + 1 < btrfs_header_nritems(leaf)) {
@@ -1027,7 +990,7 @@ delete_extent_item:
 		 * Set path->slots[0] to first slot, so that after the delete
 		 * if items are move off from our leaf to its immediate left or
 		 * right neighbor leafs, we end up with a correct and adjusted
-		 * path->slots[0] for our insertion (if replace_extent != 0).
+		 * path->slots[0] for our insertion (if args->replace_extent).
 		 */
 		path->slots[0] = del_slot;
 		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
@@ -1041,15 +1004,14 @@ delete_extent_item:
 	 * which case it unlocked our path, so check path->locks[0] matches a
 	 * write lock.
 	 */
-	if (!ret && replace_extent && leafs_visited == 1 &&
-	    (path->locks[0] == BTRFS_WRITE_LOCK_BLOCKING ||
-	     path->locks[0] == BTRFS_WRITE_LOCK) &&
+	if (!ret && args->replace_extent && leafs_visited == 1 &&
+	    path->locks[0] == BTRFS_WRITE_LOCK &&
 	    btrfs_leaf_free_space(leaf) >=
-	    sizeof(struct btrfs_item) + extent_item_size) {
+	    sizeof(struct btrfs_item) + args->extent_item_size) {
 
 		key.objectid = ino;
 		key.type = BTRFS_EXTENT_DATA_KEY;
-		key.offset = start;
+		key.offset = args->start;
 		if (!del_nr && path->slots[0] < btrfs_header_nritems(leaf)) {
 			struct btrfs_key slot_key;
 
@@ -1057,30 +1019,18 @@ delete_extent_item:
 			if (btrfs_comp_cpu_keys(&key, &slot_key) > 0)
 				path->slots[0]++;
 		}
-		setup_items_for_insert(root, path, &key, &extent_item_size, 1);
-		*key_inserted = 1;
+		setup_items_for_insert(root, path, &key,
+				       &args->extent_item_size, 1);
+		args->extent_inserted = true;
 	}
 
-	if (!replace_extent || !(*key_inserted))
+	if (!args->path)
+		btrfs_free_path(path);
+	else if (!args->extent_inserted)
 		btrfs_release_path(path);
-	if (drop_end)
-		*drop_end = found ? min(end, last_end) : end;
-	return ret;
-}
-
-int btrfs_drop_extents(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *root, struct inode *inode, u64 start,
-		       u64 end, int drop_cache)
-{
-	struct btrfs_path *path;
-	int ret;
+out:
+	args->drop_end = found ? min(args->end, last_end) : args->end;
 
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-	ret = __btrfs_drop_extents(trans, root, BTRFS_I(inode), path, start,
-				   end, NULL, drop_cache, 0, 0, NULL);
-	btrfs_free_path(path);
 	return ret;
 }
 
@@ -1615,11 +1565,84 @@ void btrfs_check_nocow_unlock(struct btrfs_inode *inode)
 	btrfs_drew_write_unlock(&inode->root->snapshot_lock);
 }
 
+static void update_time_for_write(struct inode *inode)
+{
+	struct timespec64 now;
+
+	if (IS_NOCMTIME(inode))
+		return;
+
+	now = current_time(inode);
+	if (!timespec64_equal(&inode->i_mtime, &now))
+		inode->i_mtime = now;
+
+	if (!timespec64_equal(&inode->i_ctime, &now))
+		inode->i_ctime = now;
+
+	if (IS_I_VERSION(inode))
+		inode_inc_iversion(inode);
+}
+
+static int btrfs_write_check(struct kiocb *iocb, struct iov_iter *from,
+			     size_t count)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	loff_t pos = iocb->ki_pos;
+	int ret;
+	loff_t oldsize;
+	loff_t start_pos;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		size_t nocow_bytes = count;
+
+		/* We will allocate space in case nodatacow is not set, so bail */
+		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes) <= 0)
+			return -EAGAIN;
+		/*
+		 * There are holes in the range or parts of the range that must
+		 * be COWed (shared extents, RO block groups, etc), so just bail
+		 * out.
+		 */
+		if (nocow_bytes < count)
+			return -EAGAIN;
+	}
+
+	current->backing_dev_info = inode_to_bdi(inode);
+	ret = file_remove_privs(file);
+	if (ret)
+		return ret;
+
+	/*
+	 * We reserve space for updating the inode when we reserve space for the
+	 * extent we are going to write, so we will enospc out there.  We don't
+	 * need to start yet another transaction to update the inode as we will
+	 * update the inode when we finish writing whatever data we write.
+	 */
+	update_time_for_write(inode);
+
+	start_pos = round_down(pos, fs_info->sectorsize);
+	oldsize = i_size_read(inode);
+	if (start_pos > oldsize) {
+		/* Expand hole size to cover write data, preventing empty gap */
+		loff_t end_pos = round_up(pos + count, fs_info->sectorsize);
+
+		ret = btrfs_cont_expand(BTRFS_I(inode), oldsize, end_pos);
+		if (ret) {
+			current->backing_dev_info = NULL;
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
 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;
+	loff_t pos;
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct page **pages = NULL;
@@ -1629,17 +1652,37 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 	u64 lockend;
 	size_t num_written = 0;
 	int nrptrs;
-	int ret = 0;
+	ssize_t ret;
 	bool only_release_metadata = false;
 	bool force_page_uptodate = false;
+	loff_t old_isize = i_size_read(inode);
+	unsigned int ilock_flags = 0;
 
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		ilock_flags |= BTRFS_ILOCK_TRY;
+
+	ret = btrfs_inode_lock(inode, ilock_flags);
+	if (ret < 0)
+		return ret;
+
+	ret = generic_write_checks(iocb, i);
+	if (ret <= 0)
+		goto out;
+
+	ret = btrfs_write_check(iocb, i, ret);
+	if (ret < 0)
+		goto out;
+
+	pos = iocb->ki_pos;
 	nrptrs = min(DIV_ROUND_UP(iov_iter_count(i), PAGE_SIZE),
 			PAGE_SIZE / (sizeof(struct page *)));
 	nrptrs = min(nrptrs, current->nr_dirtied_pause - current->nr_dirtied);
 	nrptrs = max(nrptrs, 8);
 	pages = kmalloc_array(nrptrs, sizeof(struct page *), GFP_KERNEL);
-	if (!pages)
-		return -ENOMEM;
+	if (!pages) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	while (iov_iter_count(i) > 0) {
 		struct extent_state *cached_state = NULL;
@@ -1648,8 +1691,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 		size_t write_bytes = min(iov_iter_count(i),
 					 nrptrs * (size_t)PAGE_SIZE -
 					 offset);
-		size_t num_pages = DIV_ROUND_UP(write_bytes + offset,
-						PAGE_SIZE);
+		size_t num_pages;
 		size_t reserve_bytes;
 		size_t dirty_pages;
 		size_t copied;
@@ -1657,8 +1699,6 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 		size_t num_sectors;
 		int extents_locked;
 
-		WARN_ON(num_pages > nrptrs);
-
 		/*
 		 * Fault pages before locking them in prepare_pages
 		 * to avoid recursive lock
@@ -1670,35 +1710,28 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 
 		only_release_metadata = false;
 		sector_offset = pos & (fs_info->sectorsize - 1);
-		reserve_bytes = round_up(write_bytes + sector_offset,
-				fs_info->sectorsize);
 
 		extent_changeset_release(data_reserved);
 		ret = btrfs_check_data_free_space(BTRFS_I(inode),
 						  &data_reserved, pos,
 						  write_bytes);
 		if (ret < 0) {
+			/*
+			 * If we don't have to COW at the offset, reserve
+			 * metadata only. write_bytes may get smaller than
+			 * requested here.
+			 */
 			if (btrfs_check_nocow_lock(BTRFS_I(inode), pos,
-						   &write_bytes) > 0) {
-				/*
-				 * For nodata cow case, no need to reserve
-				 * data space.
-				 */
+						   &write_bytes) > 0)
 				only_release_metadata = true;
-				/*
-				 * our prealloc extent may be smaller than
-				 * write_bytes, so scale down.
-				 */
-				num_pages = DIV_ROUND_UP(write_bytes + offset,
-							 PAGE_SIZE);
-				reserve_bytes = round_up(write_bytes +
-							 sector_offset,
-							 fs_info->sectorsize);
-			} else {
+			else
 				break;
-			}
 		}
 
+		num_pages = DIV_ROUND_UP(write_bytes + offset, PAGE_SIZE);
+		WARN_ON(num_pages > nrptrs);
+		reserve_bytes = round_up(write_bytes + sector_offset,
+					 fs_info->sectorsize);
 		WARN_ON(reserve_bytes == 0);
 		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode),
 				reserve_bytes);
@@ -1767,8 +1800,7 @@ again:
 
 		if (num_sectors > dirty_sectors) {
 			/* release everything except the sectors we dirtied */
-			release_bytes -= dirty_sectors <<
-						fs_info->sb->s_blocksize_bits;
+			release_bytes -= dirty_sectors << fs_info->sectorsize_bits;
 			if (only_release_metadata) {
 				btrfs_delalloc_release_metadata(BTRFS_I(inode),
 							release_bytes, true);
@@ -1787,10 +1819,9 @@ again:
 		release_bytes = round_up(copied + sector_offset,
 					fs_info->sectorsize);
 
-		if (copied > 0)
-			ret = btrfs_dirty_pages(BTRFS_I(inode), pages,
-						dirty_pages, pos, copied,
-						&cached_state);
+		ret = btrfs_dirty_pages(BTRFS_I(inode), pages,
+					dirty_pages, pos, copied,
+					&cached_state, only_release_metadata);
 
 		/*
 		 * If we have not locked the extent range, because the range's
@@ -1815,17 +1846,6 @@ again:
 		if (only_release_metadata)
 			btrfs_check_nocow_unlock(BTRFS_I(inode));
 
-		if (only_release_metadata && copied > 0) {
-			lockstart = round_down(pos,
-					       fs_info->sectorsize);
-			lockend = round_up(pos + copied,
-					   fs_info->sectorsize) - 1;
-
-			set_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
-				       lockend, EXTENT_NORESERVE, NULL,
-				       NULL, GFP_NOFS);
-		}
-
 		btrfs_drop_pages(pages, num_pages);
 
 		cond_resched();
@@ -1852,25 +1872,103 @@ again:
 	}
 
 	extent_changeset_free(data_reserved);
+	if (num_written > 0) {
+		pagecache_isize_extended(inode, old_isize, iocb->ki_pos);
+		iocb->ki_pos += num_written;
+	}
+out:
+	btrfs_inode_unlock(inode, ilock_flags);
 	return num_written ? num_written : ret;
 }
 
-static ssize_t __btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info,
+			       const struct iov_iter *iter, loff_t offset)
+{
+	const u32 blocksize_mask = fs_info->sectorsize - 1;
+
+	if (offset & blocksize_mask)
+		return -EINVAL;
+
+	if (iov_iter_alignment(iter) & blocksize_mask)
+		return -EINVAL;
+
+	return 0;
+}
+
+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);
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	loff_t pos;
-	ssize_t written;
+	ssize_t written = 0;
 	ssize_t written_buffered;
 	loff_t endbyte;
-	int err;
+	ssize_t err;
+	unsigned int ilock_flags = 0;
+	struct iomap_dio *dio = NULL;
+
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		ilock_flags |= BTRFS_ILOCK_TRY;
 
-	written = btrfs_direct_IO(iocb, from);
+	/* If the write DIO is within EOF, use a shared lock */
+	if (iocb->ki_pos + iov_iter_count(from) <= i_size_read(inode))
+		ilock_flags |= BTRFS_ILOCK_SHARED;
 
-	if (written < 0 || !iov_iter_count(from))
-		return written;
+relock:
+	err = btrfs_inode_lock(inode, ilock_flags);
+	if (err < 0)
+		return err;
+
+	err = generic_write_checks(iocb, from);
+	if (err <= 0) {
+		btrfs_inode_unlock(inode, ilock_flags);
+		return err;
+	}
+
+	err = btrfs_write_check(iocb, from, err);
+	if (err < 0) {
+		btrfs_inode_unlock(inode, ilock_flags);
+		goto out;
+	}
 
 	pos = iocb->ki_pos;
+	/*
+	 * Re-check since file size may have changed just before taking the
+	 * lock or pos may have changed because of O_APPEND in generic_write_check()
+	 */
+	if ((ilock_flags & BTRFS_ILOCK_SHARED) &&
+	    pos + iov_iter_count(from) > i_size_read(inode)) {
+		btrfs_inode_unlock(inode, ilock_flags);
+		ilock_flags &= ~BTRFS_ILOCK_SHARED;
+		goto relock;
+	}
+
+	if (check_direct_IO(fs_info, from, pos)) {
+		btrfs_inode_unlock(inode, ilock_flags);
+		goto buffered;
+	}
+
+	dio = __iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops,
+			     &btrfs_dio_ops, is_sync_kiocb(iocb));
+
+	btrfs_inode_unlock(inode, ilock_flags);
+
+	if (IS_ERR_OR_NULL(dio)) {
+		err = PTR_ERR_OR_ZERO(dio);
+		if (err < 0 && err != -ENOTBLK)
+			goto out;
+	} else {
+		written = iomap_dio_complete(dio);
+	}
+
+	if (written < 0 || !iov_iter_count(from)) {
+		err = written;
+		goto out;
+	}
+
+buffered:
+	pos = iocb->ki_pos;
 	written_buffered = btrfs_buffered_write(iocb, from);
 	if (written_buffered < 0) {
 		err = written_buffered;
@@ -1895,24 +1993,6 @@ out:
 	return written ? written : err;
 }
 
-static void update_time_for_write(struct inode *inode)
-{
-	struct timespec64 now;
-
-	if (IS_NOCMTIME(inode))
-		return;
-
-	now = current_time(inode);
-	if (!timespec64_equal(&inode->i_mtime, &now))
-		inode->i_mtime = now;
-
-	if (!timespec64_equal(&inode->i_ctime, &now))
-		inode->i_ctime = now;
-
-	if (IS_I_VERSION(inode))
-		inode_inc_iversion(inode);
-}
-
 static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 				    struct iov_iter *from)
 {
@@ -1920,148 +2000,28 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	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;
-	u64 start_pos;
-	u64 end_pos;
 	ssize_t num_written = 0;
 	const bool sync = iocb->ki_flags & IOCB_DSYNC;
-	ssize_t err;
-	loff_t pos;
-	size_t count;
-	loff_t oldsize;
-	int clean_page = 0;
-
-	if (!(iocb->ki_flags & IOCB_DIRECT) &&
-	    (iocb->ki_flags & IOCB_NOWAIT))
-		return -EOPNOTSUPP;
-
-	if (iocb->ki_flags & IOCB_NOWAIT) {
-		if (!inode_trylock(inode))
-			return -EAGAIN;
-	} else {
-		inode_lock(inode);
-	}
-
-	err = generic_write_checks(iocb, from);
-	if (err <= 0) {
-		inode_unlock(inode);
-		return err;
-	}
-
-	pos = iocb->ki_pos;
-	count = iov_iter_count(from);
-	if (iocb->ki_flags & IOCB_NOWAIT) {
-		size_t nocow_bytes = count;
-
-		/*
-		 * We will allocate space in case nodatacow is not set,
-		 * so bail
-		 */
-		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes)
-		    <= 0) {
-			inode_unlock(inode);
-			return -EAGAIN;
-		}
-		/*
-		 * There are holes in the range or parts of the range that must
-		 * be COWed (shared extents, RO block groups, etc), so just bail
-		 * out.
-		 */
-		if (nocow_bytes < count) {
-			inode_unlock(inode);
-			return -EAGAIN;
-		}
-	}
-
-	current->backing_dev_info = inode_to_bdi(inode);
-	err = file_remove_privs(file);
-	if (err) {
-		inode_unlock(inode);
-		goto out;
-	}
-
-	/*
-	 * If BTRFS flips readonly due to some impossible error
-	 * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR),
-	 * although we have opened a file as writable, we have
-	 * to stop this write operation to ensure FS consistency.
-	 */
-	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
-		inode_unlock(inode);
-		err = -EROFS;
-		goto out;
-	}
 
 	/*
-	 * We reserve space for updating the inode when we reserve space for the
-	 * extent we are going to write, so we will enospc out there.  We don't
-	 * need to start yet another transaction to update the inode as we will
-	 * update the inode when we finish writing whatever data we write.
+	 * If the fs flips readonly due to some impossible error, although we
+	 * have opened a file as writable, we have to stop this write operation
+	 * to ensure consistency.
 	 */
-	update_time_for_write(inode);
+	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
+		return -EROFS;
 
-	start_pos = round_down(pos, fs_info->sectorsize);
-	oldsize = i_size_read(inode);
-	if (start_pos > oldsize) {
-		/* Expand hole size to cover write data, preventing empty gap */
-		end_pos = round_up(pos + count,
-				   fs_info->sectorsize);
-		err = btrfs_cont_expand(inode, oldsize, end_pos);
-		if (err) {
-			inode_unlock(inode);
-			goto out;
-		}
-		if (start_pos > round_up(oldsize, fs_info->sectorsize))
-			clean_page = 1;
-	}
+	if (!(iocb->ki_flags & IOCB_DIRECT) &&
+	    (iocb->ki_flags & IOCB_NOWAIT))
+		return -EOPNOTSUPP;
 
 	if (sync)
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
-	if (iocb->ki_flags & IOCB_DIRECT) {
-		/*
-		 * 1. We must always clear IOCB_DSYNC in order to not deadlock
-		 *    in iomap, as it calls generic_write_sync() in this case.
-		 * 2. If we are async, we can call iomap_dio_complete() either
-		 *    in
-		 *
-		 *    2.1. A worker thread from the last bio completed.  In this
-		 *	   case we need to mark the btrfs_dio_data that it is
-		 *	   async in order to call generic_write_sync() properly.
-		 *	   This is handled by setting BTRFS_DIO_SYNC_STUB in the
-		 *	   current->journal_info.
-		 *    2.2  The submitter context, because all IO completed
-		 *         before we exited iomap_dio_rw().  In this case we can
-		 *         just re-set the IOCB_DSYNC on the iocb and we'll do
-		 *         the sync below.  If our ->end_io() gets called and
-		 *         current->journal_info is set, then we know we're in
-		 *         our current context and we will clear
-		 *         current->journal_info to indicate that we need to
-		 *         sync below.
-		 */
-		if (sync) {
-			ASSERT(current->journal_info == NULL);
-			iocb->ki_flags &= ~IOCB_DSYNC;
-			current->journal_info = BTRFS_DIO_SYNC_STUB;
-		}
-		num_written = __btrfs_direct_write(iocb, from);
-
-		/*
-		 * As stated above, we cleared journal_info, so we need to do
-		 * the sync ourselves.
-		 */
-		if (sync && current->journal_info == NULL)
-			iocb->ki_flags |= IOCB_DSYNC;
-		current->journal_info = NULL;
-	} else {
+	if (iocb->ki_flags & IOCB_DIRECT)
+		num_written = btrfs_direct_write(iocb, from);
+	else
 		num_written = btrfs_buffered_write(iocb, from);
-		if (num_written > 0)
-			iocb->ki_pos = pos + num_written;
-		if (clean_page)
-			pagecache_isize_extended(inode, oldsize,
-						i_size_read(inode));
-	}
-
-	inode_unlock(inode);
 
 	/*
 	 * We also have to set last_sub_trans to the current log transid,
@@ -2076,9 +2036,9 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 
 	if (sync)
 		atomic_dec(&BTRFS_I(inode)->sync_writers);
-out:
+
 	current->backing_dev_info = NULL;
-	return num_written ? num_written : err;
+	return num_written;
 }
 
 int btrfs_release_file(struct inode *inode, struct file *filp)
@@ -2173,13 +2133,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 
 	inode_lock(inode);
 
-	/*
-	 * We take the dio_sem here because the tree log stuff can race with
-	 * lockless dio writes and get an extent map logged for an extent we
-	 * never waited on.  We need it this high up for lockdep reasons.
-	 */
-	down_write(&BTRFS_I(inode)->dio_sem);
-
 	atomic_inc(&root->log_batch);
 
 	/*
@@ -2210,7 +2163,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 */
 	ret = start_ordered_ops(inode, start, end);
 	if (ret) {
-		up_write(&BTRFS_I(inode)->dio_sem);
 		inode_unlock(inode);
 		goto out;
 	}
@@ -2307,7 +2259,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 * file again, but that will end up using the synchronization
 	 * inside btrfs_sync_log to keep things safe.
 	 */
-	up_write(&BTRFS_I(inode)->dio_sem);
 	inode_unlock(inode);
 
 	if (ret != BTRFS_NO_LOG_SYNC) {
@@ -2338,7 +2289,6 @@ out:
 
 out_release_extents:
 	btrfs_release_log_ctx_extents(&ctx);
-	up_write(&BTRFS_I(inode)->dio_sem);
 	inode_unlock(inode);
 	goto out;
 }
@@ -2500,13 +2450,13 @@ out:
  *	   em->start + em->len > start)
  * When a hole extent is found, return 1 and modify start/len.
  */
-static int find_first_non_hole(struct inode *inode, u64 *start, u64 *len)
+static int find_first_non_hole(struct btrfs_inode *inode, u64 *start, u64 *len)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct extent_map *em;
 	int ret = 0;
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
+	em = btrfs_get_extent(inode, NULL, 0,
 			      round_down(*start, fs_info->sectorsize),
 			      round_up(*len, fs_info->sectorsize));
 	if (IS_ERR(em))
@@ -2566,13 +2516,14 @@ static int btrfs_punch_hole_lock_range(struct inode *inode,
 }
 
 static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans,
-				     struct inode *inode,
+				     struct btrfs_inode *inode,
 				     struct btrfs_path *path,
 				     struct btrfs_replace_extent_info *extent_info,
-				     const u64 replace_len)
+				     const u64 replace_len,
+				     const u64 bytes_to_drop)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_root *root = inode->root;
 	struct btrfs_file_extent_item *extent;
 	struct extent_buffer *leaf;
 	struct btrfs_key key;
@@ -2584,10 +2535,12 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans,
 		return 0;
 
 	if (extent_info->disk_offset == 0 &&
-	    btrfs_fs_incompat(fs_info, NO_HOLES))
+	    btrfs_fs_incompat(fs_info, NO_HOLES)) {
+		btrfs_update_inode_bytes(inode, 0, bytes_to_drop);
 		return 0;
+	}
 
-	key.objectid = btrfs_ino(BTRFS_I(inode));
+	key.objectid = btrfs_ino(inode);
 	key.type = BTRFS_EXTENT_DATA_KEY;
 	key.offset = extent_info->file_offset;
 	ret = btrfs_insert_empty_item(trans, root, path, &key,
@@ -2608,23 +2561,25 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_release_path(path);
 
-	ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode),
-			extent_info->file_offset, replace_len);
+	ret = btrfs_inode_set_file_extent_range(inode, extent_info->file_offset,
+						replace_len);
 	if (ret)
 		return ret;
 
 	/* If it's a hole, nothing more needs to be done. */
-	if (extent_info->disk_offset == 0)
+	if (extent_info->disk_offset == 0) {
+		btrfs_update_inode_bytes(inode, 0, bytes_to_drop);
 		return 0;
+	}
 
-	inode_add_bytes(inode, replace_len);
+	btrfs_update_inode_bytes(inode, replace_len, bytes_to_drop);
 
 	if (extent_info->is_new_extent && extent_info->insertions == 0) {
 		key.objectid = extent_info->disk_offset;
 		key.type = BTRFS_EXTENT_ITEM_KEY;
 		key.offset = extent_info->disk_len;
 		ret = btrfs_alloc_reserved_file_extent(trans, root,
-						       btrfs_ino(BTRFS_I(inode)),
+						       btrfs_ino(inode),
 						       extent_info->file_offset,
 						       extent_info->qgroup_reserved,
 						       &key);
@@ -2636,7 +2591,7 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans,
 				       extent_info->disk_len, 0);
 		ref_offset = extent_info->file_offset - extent_info->data_offset;
 		btrfs_init_data_ref(&ref, root->root_key.objectid,
-				    btrfs_ino(BTRFS_I(inode)), ref_offset);
+				    btrfs_ino(inode), ref_offset);
 		ret = btrfs_inc_extent_ref(trans, &ref);
 	}
 
@@ -2659,6 +2614,7 @@ int btrfs_replace_file_extents(struct inode *inode, struct btrfs_path *path,
 			   struct btrfs_replace_extent_info *extent_info,
 			   struct btrfs_trans_handle **trans_out)
 {
+	struct btrfs_drop_extents_args drop_args = { 0 };
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	u64 min_size = btrfs_calc_insert_metadata_size(fs_info, 1);
 	u64 ino_size = round_up(inode->i_size, fs_info->sectorsize);
@@ -2667,7 +2623,6 @@ int btrfs_replace_file_extents(struct inode *inode, struct btrfs_path *path,
 	struct btrfs_block_rsv *rsv;
 	unsigned int rsv_count;
 	u64 cur_offset;
-	u64 drop_end;
 	u64 len = end - start;
 	int ret = 0;
 
@@ -2706,10 +2661,16 @@ int btrfs_replace_file_extents(struct inode *inode, struct btrfs_path *path,
 	trans->block_rsv = rsv;
 
 	cur_offset = start;
+	drop_args.path = path;
+	drop_args.end = end + 1;
+	drop_args.drop_cache = true;
 	while (cur_offset < end) {
-		ret = __btrfs_drop_extents(trans, root, BTRFS_I(inode), path,
-					   cur_offset, end + 1, &drop_end,
-					   1, 0, 0, NULL);
+		drop_args.start = cur_offset;
+		ret = btrfs_dro