Merge branch 'misc-7.2' into next-fixes

18 files changed, 176 insertions, 61 deletions

diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index e55b686fe1ab..662cdd1cbdef 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -15,6 +15,7 @@
 #include "xattr.h"
 #include "acl.h"
 #include "misc.h"
+#include "btrfs_inode.h"
 
 struct posix_acl *btrfs_get_acl(struct inode *inode, int type, bool rcu)
 {
@@ -107,6 +108,9 @@ int btrfs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
 	struct inode *inode = d_inode(dentry);
 	umode_t old_mode = inode->i_mode;
 
+	if (btrfs_root_readonly(BTRFS_I(inode)->root))
+		return -EROFS;
+
 	if (type == ACL_TYPE_ACCESS && acl) {
 		ret = posix_acl_update_mode(idmap, inode,
 					    &inode->i_mode, &acl);
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 23c3eeb58dc1..1be632c742bd 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -3054,7 +3054,10 @@ void btrfs_backref_free_node(struct btrfs_backref_cache *cache,
 	if (node) {
 		ASSERT(list_empty(&node->list));
 		ASSERT(list_empty(&node->lower));
-		ASSERT(node->eb == NULL, "node->eb->start=%llu", node->eb->start);
+		ASSERT(node->eb == NULL, "node->eb->start=%llu level=%d owner=%llu",
+		       node->eb ? node->eb->start : 0,
+		       node->eb ? btrfs_header_level(node->eb) : 0,
+		       node->eb ? btrfs_header_owner(node->eb) : 0);
 		cache->nr_nodes--;
 		btrfs_put_root(node->root);
 		kfree(node);
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index ab76a5173272..8def7abb728f 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -4532,25 +4532,29 @@ static void reserve_chunk_space(struct btrfs_trans_handle *trans,
 		if (IS_ERR(bg)) {
 			ret = PTR_ERR(bg);
 		} else {
+			int activate_ret;
+
 			/*
 			 * We have a new chunk. We also need to activate it for
 			 * zoned filesystem.
 			 */
-			ret = btrfs_zoned_activate_one_bg(info, true);
-			if (ret < 0)
-				return;
-
-			/*
-			 * If we fail to add the chunk item here, we end up
-			 * trying again at phase 2 of chunk allocation, at
-			 * btrfs_create_pending_block_groups(). So ignore
-			 * any error here. An ENOSPC here could happen, due to
-			 * the cases described at do_chunk_alloc() - the system
-			 * block group we just created was just turned into RO
-			 * mode by a scrub for example, or a running discard
-			 * temporarily removed its free space entries, etc.
-			 */
-			btrfs_chunk_alloc_add_chunk_item(trans, bg);
+			activate_ret = btrfs_zoned_activate_one_bg(info, true);
+			if (activate_ret < 0) {
+				ret = activate_ret;
+			} else {
+				/*
+				 * If we fail to add the chunk item here, we end
+				 * up trying again at phase 2 of chunk allocation,
+				 * at btrfs_create_pending_block_groups(). So
+				 * ignore any error here. An ENOSPC here could
+				 * happen, due to the cases described at
+				 * do_chunk_alloc() - the system block group we
+				 * just created was just turned into RO mode by a
+				 * scrub for example, or a running discard
+				 * temporarily removed its free space entries, etc.
+				 */
+				btrfs_chunk_alloc_add_chunk_item(trans, bg);
+			}
 		}
 	}
 
diff --git a/fs/btrfs/block-rsv.c b/fs/btrfs/block-rsv.c
index 9efb3016ef11..c68a8f4b7d19 100644
--- a/fs/btrfs/block-rsv.c
+++ b/fs/btrfs/block-rsv.c
@@ -322,10 +322,25 @@ void btrfs_block_rsv_add_bytes(struct btrfs_block_rsv *block_rsv,
 void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
-	struct btrfs_space_info *sinfo = block_rsv->space_info;
+	struct btrfs_space_info *sinfo;
 	struct btrfs_root *root, *tmp;
-	u64 num_bytes = btrfs_root_used(&fs_info->tree_root->root_item);
 	unsigned int min_items = 1;
+	u64 num_bytes;
+
+	/*
+	 * A full read-only mount (rescue options) cannot start transactions,
+	 * so the global reserve is never consumed. Mark it as full and skip
+	 * the accounting.
+	 */
+	if (btrfs_is_full_ro(fs_info)) {
+		spin_lock(&block_rsv->lock);
+		block_rsv->full = true;
+		spin_unlock(&block_rsv->lock);
+		return;
+	}
+
+	sinfo = block_rsv->space_info;
+	num_bytes = btrfs_root_used(&fs_info->tree_root->root_item);
 
 	/*
 	 * The global block rsv is based on the size of the extent tree, the
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index d5d81f9546c3..7fdc6c3fd066 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -476,6 +476,8 @@ static inline bool btrfs_inode_can_compress(const struct btrfs_inode *inode)
 	if (inode->flags & BTRFS_INODE_NODATACOW ||
 	    inode->flags & BTRFS_INODE_NODATASUM)
 		return false;
+	if (btrfs_is_data_reloc_root(inode->root))
+		return false;
 	return true;
 }
 
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 274e2f0826b6..22c321719e4f 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -3288,15 +3288,6 @@ int btrfs_check_features(struct btrfs_fs_info *fs_info, bool is_rw_mount)
 	return 0;
 }
 
-static bool fs_is_full_ro(const struct btrfs_fs_info *fs_info)
-{
-	if (!sb_rdonly(fs_info->sb))
-		return false;
-	if (unlikely(fs_info->mount_opt & BTRFS_MOUNT_FULL_RO_MASK))
-		return true;
-	return false;
-}
-
 /*
  * Try to wait for any metadata readahead, and invalidate all btree folios.
  *
@@ -3462,7 +3453,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 		WRITE_ONCE(fs_info->fs_error, -EUCLEAN);
 
 	/* If the fs has any rescue options, no transaction is allowed. */
-	if (fs_is_full_ro(fs_info))
+	if (btrfs_is_full_ro(fs_info))
 		WRITE_ONCE(fs_info->fs_error, -EROFS);
 
 	/* Set up fs_info before parsing mount options */
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 7d604524e83c..de5785117a47 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2004,7 +2004,7 @@ static noinline_for_stack bool lock_extent_buffer_for_io(struct extent_buffer *e
 
 		btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
 		percpu_counter_add_batch(&fs_info->dirty_metadata_bytes,
-					 -eb->len,
+					 -(s64)eb->len,
 					 fs_info->dirty_metadata_batch);
 		ret = true;
 	} else {
@@ -3774,7 +3774,7 @@ void btrfs_clear_buffer_dirty(struct btrfs_trans_handle *trans,
 		return;
 
 	buffer_tree_clear_mark(eb, PAGECACHE_TAG_DIRTY);
-	percpu_counter_add_batch(&fs_info->dirty_metadata_bytes, -eb->len,
+	percpu_counter_add_batch(&fs_info->dirty_metadata_bytes, -(s64)eb->len,
 				 fs_info->dirty_metadata_batch);
 
 	for (int i = 0; i < num_extent_folios(eb); i++) {
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index fce9c5cc0122..6ad7b39ae358 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -866,13 +866,13 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end,
 			goto next;
 		}
 
-		flags = em->flags;
 		/*
 		 * In case we split the extent map, we want to preserve the
 		 * EXTENT_FLAG_LOGGING flag on our extent map, but we don't want
 		 * it on the new extent maps.
 		 */
-		em->flags &= ~(EXTENT_FLAG_PINNED | EXTENT_FLAG_LOGGING);
+		flags = em->flags & ~EXTENT_FLAG_LOGGING;
+		em->flags &= ~EXTENT_FLAG_PINNED;
 		modified = !list_empty(&em->list);
 
 		/*
diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h
index 5f0cfb0b5466..7ee9ec2b0efb 100644
--- a/fs/btrfs/fs.h
+++ b/fs/btrfs/fs.h
@@ -1159,6 +1159,15 @@ void __btrfs_clear_fs_compat_ro(struct btrfs_fs_info *fs_info, u64 flag,
 #define btrfs_test_opt(fs_info, opt)	((fs_info)->mount_opt & \
 					 BTRFS_MOUNT_##opt)
 
+static inline bool btrfs_is_full_ro(const struct btrfs_fs_info *fs_info)
+{
+	if (!sb_rdonly(fs_info->sb))
+		return false;
+	if (unlikely(fs_info->mount_opt & BTRFS_MOUNT_FULL_RO_MASK))
+		return true;
+	return false;
+}
+
 static inline bool btrfs_fs_closing(const struct btrfs_fs_info *fs_info)
 {
 	return unlikely(test_bit(BTRFS_FS_CLOSING_START, &fs_info->flags));
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 272598f6ae77..b446c3014b24 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -8068,7 +8068,8 @@ static int btrfs_getattr(struct mnt_idmap *idmap,
 	stat->result_mask |= STATX_SUBVOL;
 
 	spin_lock(&BTRFS_I(inode)->lock);
-	delalloc_bytes = BTRFS_I(inode)->new_delalloc_bytes;
+	delalloc_bytes = S_ISREG(inode->i_mode) ?
+			 BTRFS_I(inode)->new_delalloc_bytes : 0;
 	inode_bytes = inode_get_bytes(inode);
 	spin_unlock(&BTRFS_I(inode)->lock);
 	stat->blocks = (ALIGN(inode_bytes, blocksize) +
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 81e87bc39828..baa645e98812 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -2048,6 +2048,7 @@ static int _btrfs_ioctl_get_subvol_info(struct inode *inode,
 			ret = -ENOENT;
 			goto out;
 		}
+		ret = 0;
 	}
 
 out:
@@ -5204,7 +5205,7 @@ static int btrfs_ioctl_get_csums(struct file *file, void __user *argp)
 	struct btrfs_inode *inode = BTRFS_I(vfs_inode);
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_root *root = inode->root;
-	struct btrfs_ioctl_get_csums_args args;
+	struct btrfs_ioctl_get_csums_args args = { 0 };
 	BTRFS_PATH_AUTO_FREE(path);
 	const u64 ino = btrfs_ino(inode);
 	const u32 csum_size = fs_info->csum_size;
diff --git a/fs/btrfs/raid-stripe-tree.c b/fs/btrfs/raid-stripe-tree.c
index 454a95bf542a..b210371ce91e 100644
--- a/fs/btrfs/raid-stripe-tree.c
+++ b/fs/btrfs/raid-stripe-tree.c
@@ -414,6 +414,12 @@ int btrfs_get_raid_extent_offset(struct btrfs_fs_info *fs_info,
 	int slot;
 	int ret;
 
+	if (unlikely(!stripe_root)) {
+		btrfs_err_rl(fs_info, "missing raid stripe tree root for logical %llu",
+			     logical);
+		return -EUCLEAN;
+	}
+
 	stripe_key.objectid = logical;
 	stripe_key.type = BTRFS_RAID_STRIPE_KEY;
 	stripe_key.offset = 0;
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index 00a01b97cc0c..ffb654d36391 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -1679,8 +1679,10 @@ static void verify_bio_data_sectors(struct btrfs_raid_bio *rbio,
 			continue;
 
 		/* No csum for this sector, skip to the next sector. */
-		if (!test_bit(total_sector_nr, rbio->csum_bitmap))
+		if (!test_bit(total_sector_nr, rbio->csum_bitmap)) {
+			total_sector_nr++;
 			continue;
+		}
 
 		expected_csum = rbio->csum_buf + total_sector_nr * fs_info->csum_size;
 		btrfs_calculate_block_csum_pages(fs_info, paddrs, csum_buf);
@@ -2909,13 +2911,12 @@ static int scrub_assemble_read_bios(struct btrfs_raid_bio *rbio)
 			continue;
 
 		/*
-		 * We want to find all the sectors missing from the rbio and
-		 * read them from the disk. If sector_paddr_in_rbio() finds a sector
-		 * in the bio list we don't need to read it off the stripe.
+		 * A parity-scrub rbio carries no data in its bio list: the
+		 * only bio there is the empty completion bio added by
+		 * raid56_parity_alloc_scrub_rbio().  Every sector is read
+		 * from the stripe, so only assert that invariant here.
 		 */
-		paddrs = sector_paddrs_in_rbio(rbio, stripe, sectornr, 1);
-		if (paddrs == NULL)
-			continue;
+		ASSERT(!sector_paddrs_in_rbio(rbio, stripe, sectornr, 1));
 
 		paddrs = rbio_stripe_paddrs(rbio, stripe, sectornr);
 		/*
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 0d63d117db59..fc5c14b5adad 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -588,6 +588,7 @@ static int __add_reloc_root(struct btrfs_root *root, struct reloc_control *rc)
 		btrfs_err(fs_info,
 			    "Duplicate root found for start=%llu while inserting into relocation tree",
 			    node->bytenr);
+		kfree(node);
 		return -EEXIST;
 	}
 
@@ -890,6 +891,13 @@ static int get_new_location(struct inode *reloc_inode, u64 *new_bytenr,
 	leaf = path->nodes[0];
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
+	if (unlikely(btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE)) {
+		btrfs_print_leaf(leaf);
+		btrfs_err(fs_info,
+	"unexpected inline file extent item for data reloc inode %llu key offset %llu",
+			  btrfs_ino(BTRFS_I(reloc_inode)), bytenr);
+		return -EUCLEAN;
+	}
 
 	/*
 	 * The cluster-boundary key searched above is always written by
@@ -1518,6 +1526,17 @@ static int insert_dirty_subvol(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
+static void clear_reloc_root(struct btrfs_root *root)
+{
+	root->reloc_root = NULL;
+	/*
+	 * Need barrier to ensure clear_bit() only happens after
+	 * root->reloc_root = NULL. Pairs with have_reloc_root().
+	 */
+	smp_wmb();
+	clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
+}
+
 static int clean_dirty_subvols(struct reloc_control *rc)
 {
 	struct btrfs_root *root;
@@ -1532,13 +1551,7 @@ static int clean_dirty_subvols(struct reloc_control *rc)
 			struct btrfs_root *reloc_root = root->reloc_root;
 
 			list_del_init(&root->reloc_dirty_list);
-			root->reloc_root = NULL;
-			/*
-			 * Need barrier to ensure clear_bit() only happens after
-			 * root->reloc_root = NULL. Pairs with have_reloc_root.
-			 */
-			smp_wmb();
-			clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
+			clear_reloc_root(root);
 			if (reloc_root) {
 				/*
 				 * btrfs_drop_snapshot drops our ref we hold for
@@ -1914,21 +1927,39 @@ again:
 				goto out;
 			}
 			ret = merge_reloc_root(rc, root);
-			btrfs_put_root(root);
 			if (ret) {
-				if (list_empty(&reloc_root->root_list))
+				/*
+				 * Clear the reloc root since below we will call
+				 * free_reloc_roots(), otherwise we leave
+				 * root->reloc_root pointing to a freed reloc
+				 * root and trigger a use-after-free during
+				 * unmount or elsewhere.
+				 */
+				clear_reloc_root(root);
+				btrfs_put_root(root);
+				/*
+				 * We are adding the reloc_root to the local
+				 * reloc_roots list, so we add a ref for this
+				 * list which will be dropped below by the call
+				 * to free_reloc_roots().
+				 */
+				if (list_empty(&reloc_root->root_list)) {
 					list_add_tail(&reloc_root->root_list,
 						      &reloc_roots);
+					btrfs_grab_root(reloc_root);
+				}
+				/* Now drop the ref for root->reloc_root. */
+				btrfs_put_root(reloc_root);
 				goto out;
 			}
+			btrfs_put_root(root);
 		} else {
 			if (!IS_ERR(root)) {
 				if (root->reloc_root == reloc_root) {
-					root->reloc_root = NULL;
+					clear_reloc_root(root);
+					/* Drop the ref for root->reloc_root. */
 					btrfs_put_root(reloc_root);
 				}
-				clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE,
-					  &root->state);
 				btrfs_put_root(root);
 			}
 
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 3ae480c7474b..297704edf1b4 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -8251,7 +8251,7 @@ out:
 	}
 
 	if (sort_clone_roots) {
-		for (i = 0; i < sctx->clone_roots_cnt; i++) {
+		for (i = 0; sctx && i < sctx->clone_roots_cnt; i++) {
 			btrfs_root_dec_send_in_progress(
 					sctx->clone_roots[i].root);
 			btrfs_put_root(sctx->clone_roots[i].root);
diff --git a/fs/btrfs/subpage.c b/fs/btrfs/subpage.c
index 56060acac2e9..2a9397be8116 100644
--- a/fs/btrfs/subpage.c
+++ b/fs/btrfs/subpage.c
@@ -359,6 +359,23 @@ void btrfs_subpage_set_dirty(const struct btrfs_fs_info *fs_info,
 	folio_mark_dirty(folio);
 }
 
+static void folio_clear_tags(struct folio *folio)
+{
+	struct address_space *mapping = folio_mapping(folio);
+	XA_STATE(xas, &mapping->i_pages, folio->index);
+	unsigned long flags;
+
+	ASSERT(folio_test_locked(folio));
+	ASSERT(mapping);
+	ASSERT(mapping_use_writeback_tags(mapping));
+
+	xas_lock_irqsave(&xas, flags);
+	xas_load(&xas);
+	xas_clear_mark(&xas, PAGECACHE_TAG_DIRTY);
+	xas_clear_mark(&xas, PAGECACHE_TAG_TOWRITE);
+	xas_unlock_irqrestore(&xas, flags);
+}
+
 /*
  * Extra clear_and_test function for subpage dirty bitmap.
  *
@@ -403,7 +420,6 @@ void btrfs_subpage_set_writeback(const struct btrfs_fs_info *fs_info,
 	unsigned int start_bit = subpage_calc_start_bit(fs_info, folio,
 							writeback, start, len);
 	unsigned long flags;
-	bool keep_write;
 
 	spin_lock_irqsave(&bfs->lock, flags);
 	bitmap_set(bfs->bitmaps, start_bit, len >> fs_info->sectorsize_bits);
@@ -413,10 +429,14 @@ void btrfs_subpage_set_writeback(const struct btrfs_fs_info *fs_info,
 	 * folio. Doing so can cause WB_SYNC_ALL writepages() to overlook it,
 	 * assume writeback is complete, and exit too early — violating sync
 	 * ordering guarantees.
+	 *
+	 * Instead we manually clear the DIRTY and TOWRITE tags after the folio
+	 * is no longer dirty.
 	 */
-	keep_write = folio_test_dirty(folio);
 	if (!folio_test_writeback(folio))
-		__folio_start_writeback(folio, keep_write);
+		__folio_start_writeback(folio, true);
+	if (!folio_test_dirty(folio))
+		folio_clear_tags(folio);
 	spin_unlock_irqrestore(&bfs->lock, flags);
 }
 
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index a7d804219bec..4b2c07211c30 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -1519,12 +1519,14 @@ static int btrfs_reconfigure(struct fs_context *fc)
 	sync_filesystem(sb);
 	set_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
 
-	if (!btrfs_check_options(fs_info, &ctx->mount_opt, fc->sb_flags))
-		return -EINVAL;
+	if (!btrfs_check_options(fs_info, &ctx->mount_opt, fc->sb_flags)) {
+		ret = -EINVAL;
+		goto restore;
+	}
 
 	ret = btrfs_check_features(fs_info, !(fc->sb_flags & SB_RDONLY));
 	if (ret < 0)
-		return ret;
+		goto restore;
 
 	btrfs_ctx_to_info(fs_info, ctx);
 	btrfs_remount_begin(fs_info, old_ctx.mount_opt, fc->sb_flags);
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index 97f06dd01693..a016cb471beb 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -2138,6 +2138,16 @@ void btrfs_finish_ordered_zoned(struct btrfs_ordered_extent *ordered)
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags))
 		return;
 
+	/*
+	 * A fully truncated ordered extent wrote no data and so has
+	 * no zone append result to record.
+	 */
+	if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags) &&
+	    ordered->truncated_len == 0) {
+		ASSERT(list_empty(&ordered->csum_list));
+		return;
+	}
+
 	ASSERT(!list_empty(&ordered->csum_list));
 	sum = list_first_entry(&ordered->csum_list, struct btrfs_ordered_sum, list);
 	logical = sum->logical;
@@ -2190,7 +2200,11 @@ static bool check_bg_is_active(struct btrfs_eb_write_context *ctx,
 
 	if (fs_info->treelog_bg == block_group->start) {
 		if (!btrfs_zone_activate(block_group)) {
-			int ret_fin = btrfs_zone_finish_one_bg(fs_info);
+			int ret_fin;
+
+			btrfs_zoned_meta_io_unlock(fs_info);
+			ret_fin = btrfs_zone_finish_one_bg(fs_info);
+			btrfs_zoned_meta_io_lock(fs_info);
 
 			if (ret_fin != 1 || !btrfs_zone_activate(block_group))
 				return false;
@@ -3186,6 +3200,17 @@ int btrfs_reset_unused_block_groups(struct btrfs_space_info *space_info, u64 num
 		bg->zone_unusable = bg->length - bg->zone_capacity;
 		bg->alloc_offset = 0;
 		/*
+		 * The zone was just reset to empty, so alloc_offset went back to
+		 * the start of the zone. For metadata/system block groups the
+		 * write pointer must follow it back to the start of the zone;
+		 * otherwise it stays stale at the previous (finished) zone end,
+		 * and metadata written into the reused zone would sit behind the
+		 * write pointer, could never be written out in sequential order,
+		 * and would be stranded (pinning its folio) until unmount.
+		 */
+		if (bg->flags & (BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_SYSTEM))
+			bg->meta_write_pointer = bg->start;
+		/*
 		 * This holds because we currently reset fully used then freed
 		 * block group.
 		 */