Merge branch 'for-next-next-v6.12-20241009' into for-next-20241009

fs/btrfs/Kconfig

@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+# misc-next marker
 
 config BTRFS_FS
 	tristate "Btrfs filesystem support"
@@ -78,6 +79,32 @@ config BTRFS_ASSERT
 
 	  If unsure, say N.
 
+config BTRFS_EXPERIMENTAL
+	bool "Btrfs experimental features"
+	depends on BTRFS_FS
+	default n
+	help
+	  Enable experimental features.  These features may not be stable enough
+	  for end users.  This is meant for btrfs developers or users who wish
+	  to test the functionality and report problems.
+
+	  Current list:
+
+	  - extent map shrinker - performance problems with too frequent shrinks
+
+	  - send stream protocol v3 - fs-verity support
+
+	  - checksum offload mode - sysfs knob to affect when checksums are
+	                            calculated (at IO time, or in a thread)
+
+	  - raid-stripe-tree - additional mapping of extents to devices to
+			       support RAID1* profiles on zoned devices,
+			       RAID56 not yet supported
+
+	  - extent tree v2 - complex rework of extent tracking
+
+	  If unsure, say N.
+
 config BTRFS_FS_REF_VERIFY
 	bool "Btrfs with the ref verify tool compiled in"
 	depends on BTRFS_FS

fs/btrfs/Makefile

@@ -43,4 +43,5 @@ btrfs-$(CONFIG_FS_VERITY) += verity.o
 btrfs-$(CONFIG_BTRFS_FS_RUN_SANITY_TESTS) += tests/free-space-tests.o \
 	tests/extent-buffer-tests.o tests/btrfs-tests.o \
 	tests/extent-io-tests.o tests/inode-tests.o tests/qgroup-tests.o \
-	tests/free-space-tree-tests.o tests/extent-map-tests.o
+	tests/free-space-tree-tests.o tests/extent-map-tests.o \
+	tests/raid-stripe-tree-tests.o

fs/btrfs/backref.c

@@ -3021,9 +3021,6 @@ void btrfs_backref_init_cache(struct btrfs_fs_info *fs_info,
 	cache->rb_root = RB_ROOT;
 	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
 		INIT_LIST_HEAD(&cache->pending[i]);
-	INIT_LIST_HEAD(&cache->changed);
-	INIT_LIST_HEAD(&cache->detached);
-	INIT_LIST_HEAD(&cache->leaves);
 	INIT_LIST_HEAD(&cache->pending_edge);
 	INIT_LIST_HEAD(&cache->useless_node);
 	cache->fs_info = fs_info;
@@ -3131,29 +3128,17 @@ void btrfs_backref_drop_node(struct btrfs_backref_cache *tree,
 void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache,
 				struct btrfs_backref_node *node)
 {
-	struct btrfs_backref_node *upper;
 	struct btrfs_backref_edge *edge;
 
 	if (!node)
 		return;
 
-	BUG_ON(!node->lowest && !node->detached);
 	while (!list_empty(&node->upper)) {
 		edge = list_entry(node->upper.next, struct btrfs_backref_edge,
 				  list[LOWER]);
-		upper = edge->node[UPPER];
 		list_del(&edge->list[LOWER]);
 		list_del(&edge->list[UPPER]);
 		btrfs_backref_free_edge(cache, edge);
-
-		/*
-		 * Add the node to leaf node list if no other child block
-		 * cached.
-		 */
-		if (list_empty(&upper->lower)) {
-			list_add_tail(&upper->lower, &cache->leaves);
-			upper->lowest = 1;
-		}
 	}
 
 	btrfs_backref_drop_node(cache, node);
@@ -3165,33 +3150,13 @@ void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache,
 void btrfs_backref_release_cache(struct btrfs_backref_cache *cache)
 {
 	struct btrfs_backref_node *node;
-	int i;
-
-	while (!list_empty(&cache->detached)) {
-		node = list_entry(cache->detached.next,
-				  struct btrfs_backref_node, list);
-		btrfs_backref_cleanup_node(cache, node);
-	}
 
-	while (!list_empty(&cache->leaves)) {
-		node = list_entry(cache->leaves.next,
-				  struct btrfs_backref_node, lower);
+	while ((node = rb_entry_safe(rb_first(&cache->rb_root),
+				     struct btrfs_backref_node, rb_node)))
 		btrfs_backref_cleanup_node(cache, node);
-	}
 
-	for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
-		while (!list_empty(&cache->pending[i])) {
-			node = list_first_entry(&cache->pending[i],
-						struct btrfs_backref_node,
-						list);
-			btrfs_backref_cleanup_node(cache, node);
-		}
-	}
 	ASSERT(list_empty(&cache->pending_edge));
 	ASSERT(list_empty(&cache->useless_node));
-	ASSERT(list_empty(&cache->changed));
-	ASSERT(list_empty(&cache->detached));
-	ASSERT(RB_EMPTY_ROOT(&cache->rb_root));
 	ASSERT(!cache->nr_nodes);
 	ASSERT(!cache->nr_edges);
 }
@@ -3315,8 +3280,16 @@ static int handle_indirect_tree_backref(struct btrfs_trans_handle *trans,
 	root = btrfs_get_fs_root(fs_info, ref_key->offset, false);
 	if (IS_ERR(root))
 		return PTR_ERR(root);
-	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
-		cur->cowonly = 1;
+
+	/*
+	 * We shouldn't be using backref cache for non shareable roots, ASSERT
+	 * for developers, return -EUCLEAN for users.
+	 */
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
+		btrfs_put_root(root);
+		ASSERT(0);
+		return -EUCLEAN;
+	}
 
 	if (btrfs_root_level(&root->root_item) == cur->level) {
 		/* Tree root */
@@ -3402,8 +3375,20 @@ static int handle_indirect_tree_backref(struct btrfs_trans_handle *trans,
 				goto out;
 			}
 			upper->owner = btrfs_header_owner(eb);
-			if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
-				upper->cowonly = 1;
+
+			/*
+			 * We shouldn't be using backref cache for non shareable
+			 * roots, ASSERT for developers, return -EUCLEAN for
+			 * users.
+			 */
+			if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
+				btrfs_put_root(root);
+				btrfs_backref_free_edge(cache, edge);
+				btrfs_backref_free_node(cache, upper);
+				ASSERT(0);
+				ret = -EUCLEAN;
+				goto out;
+			}
 
 			/*
 			 * If we know the block isn't shared we can avoid
@@ -3594,15 +3579,11 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache,
 
 	ASSERT(start->checked);
 
-	/* Insert this node to cache if it's not COW-only */
-	if (!start->cowonly) {
-		rb_node = rb_simple_insert(&cache->rb_root, start->bytenr,
-					   &start->rb_node);
-		if (rb_node)
-			btrfs_backref_panic(cache->fs_info, start->bytenr,
-					    -EEXIST);
-		list_add_tail(&start->lower, &cache->leaves);
-	}
+	rb_node = rb_simple_insert(&cache->rb_root, start->bytenr,
+				   &start->rb_node);
+	if (rb_node)
+		btrfs_backref_panic(cache->fs_info, start->bytenr,
+				    -EEXIST);
 
 	/*
 	 * Use breadth first search to iterate all related edges.
@@ -3641,11 +3622,6 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache,
 		 * parents have already been linked.
 		 */
 		if (!RB_EMPTY_NODE(&upper->rb_node)) {
-			if (upper->lowest) {
-				list_del_init(&upper->lower);
-				upper->lowest = 0;
-			}
-
 			list_add_tail(&edge->list[UPPER], &upper->lower);
 			continue;
 		}
@@ -3656,23 +3632,14 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache,
 			return -EUCLEAN;
 		}
 
-		/* Sanity check, COW-only node has non-COW-only parent */
-		if (start->cowonly != upper->cowonly) {
-			ASSERT(0);
+		rb_node = rb_simple_insert(&cache->rb_root, upper->bytenr,
+					   &upper->rb_node);
+		if (rb_node) {
+			btrfs_backref_panic(cache->fs_info,
+					    upper->bytenr, -EEXIST);
 			return -EUCLEAN;
 		}
 
-		/* Only cache non-COW-only (subvolume trees) tree blocks */
-		if (!upper->cowonly) {
-			rb_node = rb_simple_insert(&cache->rb_root, upper->bytenr,
-						   &upper->rb_node);
-			if (rb_node) {
-				btrfs_backref_panic(cache->fs_info,
-						upper->bytenr, -EEXIST);
-				return -EUCLEAN;
-			}
-		}
-
 		list_add_tail(&edge->list[UPPER], &upper->lower);
 
 		/*

fs/btrfs/backref.h

@@ -318,6 +318,12 @@ struct btrfs_backref_node {
 		u64 bytenr;
 	}; /* Use rb_simple_node for search/insert */
 
+	/*
+	 * This is a sanity check, whenever we cow a block we will update
+	 * new_bytenr with it's current location, and we will check this in
+	 * various places to validate that the cache makes sense, it shouldn't
+	 * be used for anything else.
+	 */
 	u64 new_bytenr;
 	/* Objectid of tree block owner, can be not uptodate */
 	u64 owner;
@@ -335,10 +341,6 @@ struct btrfs_backref_node {
 	struct extent_buffer *eb;
 	/* Level of the tree block */
 	unsigned int level:8;
-	/* Is the block in a non-shareable tree */
-	unsigned int cowonly:1;
-	/* 1 if no child node is in the cache */
-	unsigned int lowest:1;
 	/* Is the extent buffer locked */
 	unsigned int locked:1;
 	/* Has the block been processed */
@@ -391,12 +393,6 @@ struct btrfs_backref_cache {
 	 * level blocks may not reflect the new location
 	 */
 	struct list_head pending[BTRFS_MAX_LEVEL];
-	/* List of backref nodes with no child node */
-	struct list_head leaves;
-	/* List of blocks that have been COWed in current transaction */
-	struct list_head changed;
-	/* List of detached backref node. */
-	struct list_head detached;
 
 	u64 last_trans;
 

fs/btrfs/bio.c

@@ -598,7 +598,7 @@ static bool should_async_write(struct btrfs_bio *bbio)
 {
 	bool auto_csum_mode = true;
 
-#ifdef CONFIG_BTRFS_DEBUG
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
 	struct btrfs_fs_devices *fs_devices = bbio->fs_info->fs_devices;
 	enum btrfs_offload_csum_mode csum_mode = READ_ONCE(fs_devices->offload_csum_mode);
 

fs/btrfs/block-group.c

@@ -2797,7 +2797,7 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 		 * uncompressed data size, because the compression is only done
 		 * when writeback triggered and we don't know how much space we
 		 * are actually going to need, so we reserve the uncompressed
-		 * size because the data may be uncompressible in the worst case.
+		 * size because the data may be incompressible in the worst case.
 		 */
 		if (ret == 0) {
 			bool used;

fs/btrfs/btrfs_inode.h

@@ -577,7 +577,6 @@ void btrfs_merge_delalloc_extent(struct btrfs_inode *inode, struct extent_state
 				 struct extent_state *other);
 void btrfs_split_delalloc_extent(struct btrfs_inode *inode,
 				 struct extent_state *orig, u64 split);
-void btrfs_set_range_writeback(struct btrfs_inode *inode, u64 start, u64 end);
 void btrfs_evict_inode(struct inode *inode);
 struct inode *btrfs_alloc_inode(struct super_block *sb);
 void btrfs_destroy_inode(struct inode *inode);

fs/btrfs/compression.c

@@ -1030,13 +1030,16 @@ int btrfs_compress_folios(unsigned int type_level, struct address_space *mapping
 {
 	int type = btrfs_compress_type(type_level);
 	int level = btrfs_compress_level(type_level);
+	const unsigned long orig_len = *total_out;
 	struct list_head *workspace;
 	int ret;
 
 	level = btrfs_compress_set_level(type, level);
 	workspace = get_workspace(type, level);
 	ret = compression_compress_pages(type, workspace, mapping, start, folios,
 					 out_folios, total_in, total_out);
+	/* The total read-in bytes should be no larger than the input. */
+	ASSERT(*total_in <= orig_len);
 	put_workspace(type, workspace);
 	return ret;
 }

fs/btrfs/delayed-ref.c

@@ -830,7 +830,6 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref,
 			qrecord->data_rsv = reserved;
 			qrecord->data_rsv_refroot = generic_ref->ref_root;
 		}
-		qrecord->bytenr = generic_ref->bytenr;
 		qrecord->num_bytes = generic_ref->num_bytes;
 		qrecord->old_roots = NULL;
 	}
@@ -849,6 +848,7 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans,
 		     struct btrfs_qgroup_extent_record *qrecord,
 		     int action, bool *qrecord_inserted_ret)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_ref_head *existing;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	bool qrecord_inserted = false;
@@ -859,11 +859,12 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans,
 	if (qrecord) {
 		int ret;
 
-		ret = btrfs_qgroup_trace_extent_nolock(trans->fs_info,
-						       delayed_refs, qrecord);
+		ret = btrfs_qgroup_trace_extent_nolock(fs_info, delayed_refs, qrecord,
+						       head_ref->bytenr);
 		if (ret) {
 			/* Clean up if insertion fails or item exists. */
-			xa_release(&delayed_refs->dirty_extents, qrecord->bytenr);
+			xa_release(&delayed_refs->dirty_extents,
+				   head_ref->bytenr >> fs_info->sectorsize_bits);
 			/* Caller responsible for freeing qrecord on error. */
 			if (ret < 0)
 				return ERR_PTR(ret);
@@ -873,7 +874,7 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans,
 		}
 	}
 
-	trace_add_delayed_ref_head(trans->fs_info, head_ref, action);
+	trace_add_delayed_ref_head(fs_info, head_ref, action);
 
 	existing = htree_insert(&delayed_refs->href_root,
 				&head_ref->href_node);
@@ -895,8 +896,7 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans,
 		if (head_ref->is_data && head_ref->ref_mod < 0) {
 			delayed_refs->pending_csums += head_ref->num_bytes;
 			trans->delayed_ref_csum_deletions +=
-				btrfs_csum_bytes_to_leaves(trans->fs_info,
-							   head_ref->num_bytes);
+				btrfs_csum_bytes_to_leaves(fs_info, head_ref->num_bytes);
 		}
 		delayed_refs->num_heads++;
 		delayed_refs->num_heads_ready++;
@@ -1030,7 +1030,8 @@ static int add_delayed_ref(struct btrfs_trans_handle *trans,
 			goto free_head_ref;
 		}
 		if (xa_reserve(&trans->transaction->delayed_refs.dirty_extents,
-			       generic_ref->bytenr, GFP_NOFS)) {
+			       generic_ref->bytenr >> fs_info->sectorsize_bits,
+			       GFP_NOFS)) {
 			ret = -ENOMEM;
 			goto free_record;
 		}
@@ -1073,7 +1074,7 @@ static int add_delayed_ref(struct btrfs_trans_handle *trans,
 		kmem_cache_free(btrfs_delayed_ref_node_cachep, node);
 
 	if (qrecord_inserted)
-		return btrfs_qgroup_trace_extent_post(trans, record);
+		return btrfs_qgroup_trace_extent_post(trans, record, head_ref->bytenr);
 	return 0;
 
 free_record:

fs/btrfs/delayed-ref.h

@@ -202,7 +202,15 @@ struct btrfs_delayed_ref_root {
 	/* head ref rbtree */
 	struct rb_root_cached href_root;
 
-	/* Track dirty extent records. */
+	/*
+	 * Track dirty extent records.
+	 * The keys correspond to the logical address of the extent ("bytenr")
+	 * right shifted by fs_info->sectorsize_bits. This is both to get a more
+	 * dense index space (optimizes xarray structure) and because indexes in
+	 * xarrays are of "unsigned long" type, meaning they are 32 bits wide on
+	 * 32 bits platforms, limiting the extent range to 4G which is too low
+	 * and makes it unusable (truncated index values) on 32 bits platforms.
+	 */
 	struct xarray dirty_extents;
 
 	/* this spin lock protects the rbtree and the entries inside */