Merge tag 'for-6.10-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull more btrfs updates from David Sterba:
 "A few more updates, mostly stability fixes or user visible changes:

   - fix race in zoned mode during device replace that can lead to
     use-after-free

   - update return codes and lower message levels for quota rescan where
     it's causing false alerts

   - fix unexpected qgroup id reuse under some conditions

   - fix condition when looking up extent refs

   - add option norecovery (removed in 6.8), the intended replacements
     haven't been used and some aplications still rely on the old one

   - build warning fixes"

* tag 'for-6.10-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: re-introduce 'norecovery' mount option
  btrfs: fix end of tree detection when searching for data extent ref
  btrfs: scrub: initialize ret in scrub_simple_mirror() to fix compilation warning
  btrfs: zoned: fix use-after-free due to race with dev replace
  btrfs: qgroup: fix qgroup id collision across mounts
  btrfs: qgroup: update rescan message levels and error codes

Author: torvalds

fs/btrfs/extent-tree.c

@@ -477,7 +477,7 @@ static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans,
 		if (path->slots[0] >= nritems) {
 			ret = btrfs_next_leaf(root, path);
 			if (ret) {
-				if (ret > 1)
+				if (ret > 0)
 					return -ENOENT;
 				return ret;
 			}

fs/btrfs/qgroup.c

@@ -468,13 +468,33 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
 		}
 		if (!qgroup) {
 			struct btrfs_qgroup *prealloc;
+			struct btrfs_root *tree_root = fs_info->tree_root;
 
 			prealloc = kzalloc(sizeof(*prealloc), GFP_KERNEL);
 			if (!prealloc) {
 				ret = -ENOMEM;
 				goto out;
 			}
 			qgroup = add_qgroup_rb(fs_info, prealloc, found_key.offset);
+			/*
+			 * If a qgroup exists for a subvolume ID, it is possible
+			 * that subvolume has been deleted, in which case
+			 * re-using that ID would lead to incorrect accounting.
+			 *
+			 * Ensure that we skip any such subvol ids.
+			 *
+			 * We don't need to lock because this is only called
+			 * during mount before we start doing things like creating
+			 * subvolumes.
+			 */
+			if (is_fstree(qgroup->qgroupid) &&
+			    qgroup->qgroupid > tree_root->free_objectid)
+				/*
+				 * Don't need to check against BTRFS_LAST_FREE_OBJECTID,
+				 * as it will get checked on the next call to
+				 * btrfs_get_free_objectid.
+				 */
+				tree_root->free_objectid = qgroup->qgroupid + 1;
 		}
 		ret = btrfs_sysfs_add_one_qgroup(fs_info, qgroup);
 		if (ret < 0)
@@ -3820,14 +3840,14 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 		/* we're resuming qgroup rescan at mount time */
 		if (!(fs_info->qgroup_flags &
 		      BTRFS_QGROUP_STATUS_FLAG_RESCAN)) {
-			btrfs_warn(fs_info,
+			btrfs_debug(fs_info,
 			"qgroup rescan init failed, qgroup rescan is not queued");
 			ret = -EINVAL;
 		} else if (!(fs_info->qgroup_flags &
 			     BTRFS_QGROUP_STATUS_FLAG_ON)) {
-			btrfs_warn(fs_info,
+			btrfs_debug(fs_info,
 			"qgroup rescan init failed, qgroup is not enabled");
-			ret = -EINVAL;
+			ret = -ENOTCONN;
 		}
 
 		if (ret)
@@ -3838,14 +3858,12 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 
 	if (init_flags) {
 		if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
-			btrfs_warn(fs_info,
-				   "qgroup rescan is already in progress");
 			ret = -EINPROGRESS;
 		} else if (!(fs_info->qgroup_flags &
 			     BTRFS_QGROUP_STATUS_FLAG_ON)) {
-			btrfs_warn(fs_info,
+			btrfs_debug(fs_info,
 			"qgroup rescan init failed, qgroup is not enabled");
-			ret = -EINVAL;
+			ret = -ENOTCONN;
 		} else if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_DISABLED) {
 			/* Quota disable is in progress */
 			ret = -EBUSY;

fs/btrfs/scrub.c

@@ -2100,7 +2100,7 @@ static int scrub_simple_mirror(struct scrub_ctx *sctx,
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	const u64 logical_end = logical_start + logical_length;
 	u64 cur_logical = logical_start;
-	int ret;
+	int ret = 0;
 
 	/* The range must be inside the bg */
 	ASSERT(logical_start >= bg->start && logical_end <= bg->start + bg->length);

fs/btrfs/super.c

@@ -119,6 +119,7 @@ enum {
 	Opt_thread_pool,
 	Opt_treelog,
 	Opt_user_subvol_rm_allowed,
+	Opt_norecovery,
 
 	/* Rescue options */
 	Opt_rescue,
@@ -245,6 +246,8 @@ static const struct fs_parameter_spec btrfs_fs_parameters[] = {
 	__fsparam(NULL, "nologreplay", Opt_nologreplay, fs_param_deprecated, NULL),
 	/* Deprecated, with alias rescue=usebackuproot */
 	__fsparam(NULL, "usebackuproot", Opt_usebackuproot, fs_param_deprecated, NULL),
+	/* For compatibility only, alias for "rescue=nologreplay". */
+	fsparam_flag("norecovery", Opt_norecovery),
 
 	/* Debugging options. */
 	fsparam_flag_no("enospc_debug", Opt_enospc_debug),
@@ -438,6 +441,11 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		"'nologreplay' is deprecated, use 'rescue=nologreplay' instead");
 		btrfs_set_opt(ctx->mount_opt, NOLOGREPLAY);
 		break;
+	case Opt_norecovery:
+		btrfs_info(NULL,
+"'norecovery' is for compatibility only, recommended to use 'rescue=nologreplay'");
+		btrfs_set_opt(ctx->mount_opt, NOLOGREPLAY);
+		break;
 	case Opt_flushoncommit:
 		if (result.negated)
 			btrfs_clear_opt(ctx->mount_opt, FLUSHONCOMMIT);

fs/btrfs/zoned.c

@@ -1290,15 +1290,19 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 				struct btrfs_chunk_map *map)
 {
 	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
-	struct btrfs_device *device = map->stripes[zone_idx].dev;
+	struct btrfs_device *device;
 	int dev_replace_is_ongoing = 0;
 	unsigned int nofs_flag;
 	struct blk_zone zone;
 	int ret;
 
 	info->physical = map->stripes[zone_idx].physical;
 
+	down_read(&dev_replace->rwsem);
+	device = map->stripes[zone_idx].dev;
+
 	if (!device->bdev) {
+		up_read(&dev_replace->rwsem);
 		info->alloc_offset = WP_MISSING_DEV;
 		return 0;
 	}
@@ -1308,18 +1312,17 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 		__set_bit(zone_idx, active);
 
 	if (!btrfs_dev_is_sequential(device, info->physical)) {
+		up_read(&dev_replace->rwsem);
 		info->alloc_offset = WP_CONVENTIONAL;
 		return 0;
 	}
 
 	/* This zone will be used for allocation, so mark this zone non-empty. */
 	btrfs_dev_clear_zone_empty(device, info->physical);
 
-	down_read(&dev_replace->rwsem);
 	dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(dev_replace);
 	if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL)
 		btrfs_dev_clear_zone_empty(dev_replace->tgtdev, info->physical);
-	up_read(&dev_replace->rwsem);
 
 	/*
 	 * The group is mapped to a sequential zone. Get the zone write pointer
@@ -1330,6 +1333,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 	ret = btrfs_get_dev_zone(device, info->physical, &zone);
 	memalloc_nofs_restore(nofs_flag);
 	if (ret) {
+		up_read(&dev_replace->rwsem);
 		if (ret != -EIO && ret != -EOPNOTSUPP)
 			return ret;
 		info->alloc_offset = WP_MISSING_DEV;
@@ -1341,6 +1345,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 		"zoned: unexpected conventional zone %llu on device %s (devid %llu)",
 			zone.start << SECTOR_SHIFT, rcu_str_deref(device->name),
 			device->devid);
+		up_read(&dev_replace->rwsem);
 		return -EIO;
 	}
 
@@ -1368,6 +1373,8 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 		break;
 	}
 
+	up_read(&dev_replace->rwsem);
+
 	return 0;
 }