Merge tag 'for-6.17-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:

 - fix delayed inode tracking in xarray, eviction can race with
   insertion and leave behind a disconnected inode

 - on systems with large page (64K) and small block size (4K) fix
   compression read that can return partially filled folio

 - slightly relax compression option format for backward compatibility,
   allow to specify level for LZO although there's only one

 - fix simple quota accounting of compressed extents

 - validate minimum device size in 'device add'

 - update maintainers' entry

* tag 'for-6.17-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: don't allow adding block device of less than 1 MB
  MAINTAINERS: update btrfs entry
  btrfs: fix subvolume deletion lockup caused by inodes xarray race
  btrfs: fix corruption reading compressed range when block size is smaller than page size
  btrfs: accept and ignore compression level for lzo
  btrfs: fix squota compressed stats leak

Author: torvalds

MAINTAINERS

@@ -5258,7 +5258,6 @@ F:	drivers/gpio/gpio-bt8xx.c
 
 BTRFS FILE SYSTEM
 M:	Chris Mason <[email protected]>
-M:	Josef Bacik <[email protected]>
 M:	David Sterba <[email protected]>
 L:	[email protected]
 S:	Maintained

fs/btrfs/extent_io.c

@@ -111,6 +111,24 @@ struct btrfs_bio_ctrl {
 	 */
 	unsigned long submit_bitmap;
 	struct readahead_control *ractl;
+
+	/*
+	 * The start offset of the last used extent map by a read operation.
+	 *
+	 * This is for proper compressed read merge.
+	 * U64_MAX means we are starting the read and have made no progress yet.
+	 *
+	 * The current btrfs_bio_is_contig() only uses disk_bytenr as
+	 * the condition to check if the read can be merged with previous
+	 * bio, which is not correct. E.g. two file extents pointing to the
+	 * same extent but with different offset.
+	 *
+	 * So here we need to do extra checks to only merge reads that are
+	 * covered by the same extent map.
+	 * Just extent_map::start will be enough, as they are unique
+	 * inside the same inode.
+	 */
+	u64 last_em_start;
 };
 
 static void submit_one_bio(struct btrfs_bio_ctrl *bio_ctrl)
@@ -909,7 +927,7 @@ static void btrfs_readahead_expand(struct readahead_control *ractl,
  * return 0 on success, otherwise return error
  */
 static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached,
-		      struct btrfs_bio_ctrl *bio_ctrl, u64 *prev_em_start)
+			     struct btrfs_bio_ctrl *bio_ctrl)
 {
 	struct inode *inode = folio->mapping->host;
 	struct btrfs_fs_info *fs_info = inode_to_fs_info(inode);
@@ -1019,12 +1037,11 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached,
 		 * non-optimal behavior (submitting 2 bios for the same extent).
 		 */
 		if (compress_type != BTRFS_COMPRESS_NONE &&
-		    prev_em_start && *prev_em_start != (u64)-1 &&
-		    *prev_em_start != em->start)
+		    bio_ctrl->last_em_start != U64_MAX &&
+		    bio_ctrl->last_em_start != em->start)
 			force_bio_submit = true;
 
-		if (prev_em_start)
-			*prev_em_start = em->start;
+		bio_ctrl->last_em_start = em->start;
 
 		btrfs_free_extent_map(em);
 		em = NULL;
@@ -1238,12 +1255,15 @@ int btrfs_read_folio(struct file *file, struct folio *folio)
 	const u64 start = folio_pos(folio);
 	const u64 end = start + folio_size(folio) - 1;
 	struct extent_state *cached_state = NULL;
-	struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ };
+	struct btrfs_bio_ctrl bio_ctrl = {
+		.opf = REQ_OP_READ,
+		.last_em_start = U64_MAX,
+	};
 	struct extent_map *em_cached = NULL;
 	int ret;
 
 	lock_extents_for_read(inode, start, end, &cached_state);
-	ret = btrfs_do_readpage(folio, &em_cached, &bio_ctrl, NULL);
+	ret = btrfs_do_readpage(folio, &em_cached, &bio_ctrl);
 	btrfs_unlock_extent(&inode->io_tree, start, end, &cached_state);
 
 	btrfs_free_extent_map(em_cached);
@@ -2583,20 +2603,20 @@ void btrfs_readahead(struct readahead_control *rac)
 {
 	struct btrfs_bio_ctrl bio_ctrl = {
 		.opf = REQ_OP_READ | REQ_RAHEAD,
-		.ractl = rac
+		.ractl = rac,
+		.last_em_start = U64_MAX,
 	};
 	struct folio *folio;
 	struct btrfs_inode *inode = BTRFS_I(rac->mapping->host);
 	const u64 start = readahead_pos(rac);
 	const u64 end = start + readahead_length(rac) - 1;
 	struct extent_state *cached_state = NULL;
 	struct extent_map *em_cached = NULL;
-	u64 prev_em_start = (u64)-1;
 
 	lock_extents_for_read(inode, start, end, &cached_state);
 
 	while ((folio = readahead_folio(rac)) != NULL)
-		btrfs_do_readpage(folio, &em_cached, &bio_ctrl, &prev_em_start);
+		btrfs_do_readpage(folio, &em_cached, &bio_ctrl);
 
 	btrfs_unlock_extent(&inode->io_tree, start, end, &cached_state);
 

fs/btrfs/inode.c

@@ -5696,7 +5696,17 @@ static void btrfs_del_inode_from_root(struct btrfs_inode *inode)
 	bool empty = false;
 
 	xa_lock(&root->inodes);
-	entry = __xa_erase(&root->inodes, btrfs_ino(inode));
+	/*
+	 * This btrfs_inode is being freed and has already been unhashed at this
+	 * point. It's possible that another btrfs_inode has already been
+	 * allocated for the same inode and inserted itself into the root, so
+	 * don't delete it in that case.
+	 *
+	 * Note that this shouldn't need to allocate memory, so the gfp flags
+	 * don't really matter.
+	 */
+	entry = __xa_cmpxchg(&root->inodes, btrfs_ino(inode), inode, NULL,
+			     GFP_ATOMIC);
 	if (entry == inode)
 		empty = xa_empty(&root->inodes);
 	xa_unlock(&root->inodes);

fs/btrfs/qgroup.c

@@ -1455,6 +1455,7 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info, u64 ref_root,
 	struct btrfs_qgroup *qgroup;
 	LIST_HEAD(qgroup_list);
 	u64 num_bytes = src->excl;
+	u64 num_bytes_cmpr = src->excl_cmpr;
 	int ret = 0;
 
 	qgroup = find_qgroup_rb(fs_info, ref_root);
@@ -1466,11 +1467,12 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info, u64 ref_root,
 		struct btrfs_qgroup_list *glist;
 
 		qgroup->rfer += sign * num_bytes;
-		qgroup->rfer_cmpr += sign * num_bytes;
+		qgroup->rfer_cmpr += sign * num_bytes_cmpr;
 
 		WARN_ON(sign < 0 && qgroup->excl < num_bytes);
+		WARN_ON(sign < 0 && qgroup->excl_cmpr < num_bytes_cmpr);
 		qgroup->excl += sign * num_bytes;
-		qgroup->excl_cmpr += sign * num_bytes;
+		qgroup->excl_cmpr += sign * num_bytes_cmpr;
 
 		if (sign > 0)
 			qgroup_rsv_add_by_qgroup(fs_info, qgroup, src);

fs/btrfs/super.c

@@ -299,9 +299,12 @@ static int btrfs_parse_compress(struct btrfs_fs_context *ctx,
 		btrfs_set_opt(ctx->mount_opt, COMPRESS);
 		btrfs_clear_opt(ctx->mount_opt, NODATACOW);
 		btrfs_clear_opt(ctx->mount_opt, NODATASUM);
-	} else if (btrfs_match_compress_type(string, "lzo", false)) {
+	} else if (btrfs_match_compress_type(string, "lzo", true)) {
 		ctx->compress_type = BTRFS_COMPRESS_LZO;
-		ctx->compress_level = 0;
+		ctx->compress_level = btrfs_compress_str2level(BTRFS_COMPRESS_LZO,
+							       string + 3);
+		if (string[3] == ':' && string[4])
+			btrfs_warn(NULL, "Compression level ignored for LZO");
 		btrfs_set_opt(ctx->mount_opt, COMPRESS);
 		btrfs_clear_opt(ctx->mount_opt, NODATACOW);
 		btrfs_clear_opt(ctx->mount_opt, NODATASUM);
@@ -1079,7 +1082,7 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 			seq_printf(seq, ",compress-force=%s", compress_type);
 		else
 			seq_printf(seq, ",compress=%s", compress_type);
-		if (info->compress_level)
+		if (info->compress_level && info->compress_type != BTRFS_COMPRESS_LZO)
 			seq_printf(seq, ":%d", info->compress_level);
 	}
 	if (btrfs_test_opt(info, NOSSD))

fs/btrfs/volumes.c

@@ -2722,6 +2722,11 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		goto error;
 	}
 
+	if (bdev_nr_bytes(file_bdev(bdev_file)) <= BTRFS_DEVICE_RANGE_RESERVED) {
+		ret = -EINVAL;
+		goto error;
+	}
+
 	if (fs_devices->seeding) {
 		seeding_dev = true;
 		down_write(&sb->s_umount);