1 files changed, 119 insertions, 30 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index e6eb20987351..0117d867ecf8 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2271,13 +2271,127 @@ static blk_status_t btrfs_submit_bio_start(struct inode *inode, struct bio *bio,
 	return btrfs_csum_one_bio(BTRFS_I(inode), bio, 0, 0);
 }
 
+/*
+ * Split an extent_map at [start, start + len]
+ *
+ * This function is intended to be used only for extract_ordered_extent().
+ */
+static int split_zoned_em(struct btrfs_inode *inode, u64 start, u64 len,
+			  u64 pre, u64 post)
+{
+	struct extent_map_tree *em_tree = &inode->extent_tree;
+	struct extent_map *em;
+	struct extent_map *split_pre = NULL;
+	struct extent_map *split_mid = NULL;
+	struct extent_map *split_post = NULL;
+	int ret = 0;
+	int modified;
+	unsigned long flags;
+
+	/* Sanity check */
+	if (pre == 0 && post == 0)
+		return 0;
+
+	split_pre = alloc_extent_map();
+	if (pre)
+		split_mid = alloc_extent_map();
+	if (post)
+		split_post = alloc_extent_map();
+	if (!split_pre || (pre && !split_mid) || (post && !split_post)) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ASSERT(pre + post < len);
+
+	lock_extent(&inode->io_tree, start, start + len - 1);
+	write_lock(&em_tree->lock);
+	em = lookup_extent_mapping(em_tree, start, len);
+	if (!em) {
+		ret = -EIO;
+		goto out_unlock;
+	}
+
+	ASSERT(em->len == len);
+	ASSERT(!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags));
+	ASSERT(em->block_start < EXTENT_MAP_LAST_BYTE);
+
+	flags = em->flags;
+	clear_bit(EXTENT_FLAG_PINNED, &em->flags);
+	clear_bit(EXTENT_FLAG_LOGGING, &flags);
+	modified = !list_empty(&em->list);
+
+	/* First, replace the em with a new extent_map starting from * em->start */
+	split_pre->start = em->start;
+	split_pre->len = (pre ? pre : em->len - post);
+	split_pre->orig_start = split_pre->start;
+	split_pre->block_start = em->block_start;
+	split_pre->block_len = split_pre->len;
+	split_pre->orig_block_len = split_pre->block_len;
+	split_pre->ram_bytes = split_pre->len;
+	split_pre->flags = flags;
+	split_pre->compress_type = em->compress_type;
+	split_pre->generation = em->generation;
+
+	replace_extent_mapping(em_tree, em, split_pre, modified);
+
+	/*
+	 * Now we only have an extent_map at:
+	 *     [em->start, em->start + pre] if pre != 0
+	 *     [em->start, em->start + em->len - post] if pre == 0
+	 */
+
+	if (pre) {
+		/* Insert the middle extent_map */
+		split_mid->start = em->start + pre;
+		split_mid->len = em->len - pre - post;
+		split_mid->orig_start = split_mid->start;
+		split_mid->block_start = em->block_start + pre;
+		split_mid->block_len = split_mid->len;
+		split_mid->orig_block_len = split_mid->block_len;
+		split_mid->ram_bytes = split_mid->len;
+		split_mid->flags = flags;
+		split_mid->compress_type = em->compress_type;
+		split_mid->generation = em->generation;
+		add_extent_mapping(em_tree, split_mid, modified);
+	}
+
+	if (post) {
+		split_post->start = em->start + em->len - post;
+		split_post->len = post;
+		split_post->orig_start = split_post->start;
+		split_post->block_start = em->block_start + em->len - post;
+		split_post->block_len = split_post->len;
+		split_post->orig_block_len = split_post->block_len;
+		split_post->ram_bytes = split_post->len;
+		split_post->flags = flags;
+		split_post->compress_type = em->compress_type;
+		split_post->generation = em->generation;
+		add_extent_mapping(em_tree, split_post, modified);
+	}
+
+	/* Once for us */
+	free_extent_map(em);
+	/* Once for the tree */
+	free_extent_map(em);
+
+out_unlock:
+	write_unlock(&em_tree->lock);
+	unlock_extent(&inode->io_tree, start, start + len - 1);
+out:
+	free_extent_map(split_pre);
+	free_extent_map(split_mid);
+	free_extent_map(split_post);
+
+	return ret;
+}
+
 static blk_status_t extract_ordered_extent(struct btrfs_inode *inode,
 					   struct bio *bio, loff_t file_offset)
 {
 	struct btrfs_ordered_extent *ordered;
-	struct extent_map *em = NULL, *em_new = NULL;
-	struct extent_map_tree *em_tree = &inode->extent_tree;
 	u64 start = (u64)bio->bi_iter.bi_sector << SECTOR_SHIFT;
+	u64 file_len;
 	u64 len = bio->bi_iter.bi_size;
 	u64 end = start + len;
 	u64 ordered_end;
@@ -2317,41 +2431,16 @@ static blk_status_t extract_ordered_extent(struct btrfs_inode *inode,
 		goto out;
 	}
 
+	file_len = ordered->num_bytes;
 	pre = start - ordered->disk_bytenr;
 	post = ordered_end - end;
 
 	ret = btrfs_split_ordered_extent(ordered, pre, post);
 	if (ret)
 		goto out;
-
-	read_lock(&em_tree->lock);
-	em = lookup_extent_mapping(em_tree, ordered->file_offset, len);
-	if (!em) {
-		read_unlock(&em_tree->lock);
-		ret = -EIO;
-		goto out;
-	}
-	read_unlock(&em_tree->lock);
-
-	ASSERT(!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags));
-	/*
-	 * We cannot reuse em_new here but have to create a new one, as
-	 * unpin_extent_cache() expects the start of the extent map to be the
-	 * logical offset of the file, which does not hold true anymore after
-	 * splitting.
-	 */
-	em_new = create_io_em(inode, em->start + pre, len,
-			      em->start + pre, em->block_start + pre, len,
-			      len, len, BTRFS_COMPRESS_NONE,
-			      BTRFS_ORDERED_REGULAR);
-	if (IS_ERR(em_new)) {
-		ret = PTR_ERR(em_new);
-		goto out;
-	}
-	free_extent_map(em_new);
+	ret = split_zoned_em(inode, file_offset, file_len, pre, post);
 
 out:
-	free_extent_map(em);
 	btrfs_put_ordered_extent(ordered);
 
 	return errno_to_blk_status(ret);
@@ -2903,7 +2992,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		goto out;
 	}
 
-	if (ordered_extent->disk)
+	if (ordered_extent->bdev)
 		btrfs_rewrite_logical_zoned(ordered_extent);
 
 	btrfs_free_io_failure_record(inode, start, end);