Merge branch 'for-4.15' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "There are some new user features and the usual load of invisible
  enhancements or cleanups.

  New features:

   - extend mount options to specify zlib compression level, -o
     compress=zlib:9

   - v2 of ioctl "extent to inode mapping", addressing a usecase where
     we want to retrieve more but inaccurate results and do the
     postprocessing in userspace, aiding defragmentation or
     deduplication tools

   - populate compression heuristics logic, do data sampling and try to
     guess compressibility by: looking for repeated patterns, counting
     unique byte values and distribution, calculating Shannon entropy;
     this will need more benchmarking and possibly fine tuning, but the
     base should be good enough

   - enable indexing for btrfs as lower filesystem in overlayfs

   - speedup page cache readahead during send on large files

  Internal enhancements:

   - more sanity checks of b-tree items when reading them from disk

   - more EINVAL/EUCLEAN fixups, missing BLK_STS_* conversion, other
     errno or error handling fixes

   - remove some homegrown IO-related logic, that's been obsoleted by
     core block layer changes (batching, plug/unplug, own counters)

   - add ref-verify, optional debugging feature to verify extent
     reference accounting

   - simplify code handling outstanding extents, make it more clear
     where and how the accounting is done

   - make delalloc reservations per-inode, simplify the code and make
     the logic more straightforward

   - extensive cleanup of delayed refs code

  Notable fixes:

   - fix send ioctl on 32bit with 64bit kernel"

* 'for-4.15' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (102 commits)
  btrfs: Fix bug for misused dev_t when lookup in dev state hash table.
  Btrfs: heuristic: add Shannon entropy calculation
  Btrfs: heuristic: add byte core set calculation
  Btrfs: heuristic: add byte set calculation
  Btrfs: heuristic: add detection of repeated data patterns
  Btrfs: heuristic: implement sampling logic
  Btrfs: heuristic: add bucket and sample counters and other defines
  Btrfs: compression: separate heuristic/compression workspaces
  btrfs: move btrfs_truncate_block out of trans handle
  btrfs: don't call btrfs_start_delalloc_roots in flushoncommit
  btrfs: track refs in a rb_tree instead of a list
  btrfs: add a comp_refs() helper
  btrfs: switch args for comp_*_refs
  btrfs: make the delalloc block rsv per inode
  btrfs: add tracepoints for outstanding extents mods
  Btrfs: rework outstanding_extents
  btrfs: increase output size for LOGICAL_INO_V2 ioctl
  btrfs: add a flags argument to LOGICAL_INO and call it LOGICAL_INO_V2
  btrfs: add a flag to iterate_inodes_from_logical to find all extent refs for uncompressed extents
  btrfs: send: remove unused code
  ...

1 files changed, 19 insertions, 35 deletions

diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h
index ce88e4ac5276..a43af432f859 100644
--- a/fs/btrfs/delayed-ref.h
+++ b/fs/btrfs/delayed-ref.h
@@ -26,18 +26,8 @@
 #define BTRFS_ADD_DELAYED_EXTENT 3 /* record a full extent allocation */
 #define BTRFS_UPDATE_DELAYED_HEAD 4 /* not changing ref count on head ref */
 
-/*
- * XXX: Qu: I really hate the design that ref_head and tree/data ref shares the
- * same ref_node structure.
- * Ref_head is in a higher logic level than tree/data ref, and duplicated
- * bytenr/num_bytes in ref_node is really a waste or memory, they should be
- * referred from ref_head.
- * This gets more disgusting after we use list to store tree/data ref in
- * ref_head. Must clean this mess up later.
- */
 struct btrfs_delayed_ref_node {
-	/*data/tree ref use list, stored in ref_head->ref_list. */
-	struct list_head list;
+	struct rb_node ref_node;
 	/*
 	 * If action is BTRFS_ADD_DELAYED_REF, also link this node to
 	 * ref_head->ref_add_list, then we do not need to iterate the
@@ -91,8 +81,9 @@ struct btrfs_delayed_extent_op {
  * reference count modifications we've queued up.
  */
 struct btrfs_delayed_ref_head {
-	struct btrfs_delayed_ref_node node;
-
+	u64 bytenr;
+	u64 num_bytes;
+	refcount_t refs;
 	/*
 	 * the mutex is held while running the refs, and it is also
 	 * held when checking the sum of reference modifications.
@@ -100,7 +91,7 @@ struct btrfs_delayed_ref_head {
 	struct mutex mutex;
 
 	spinlock_t lock;
-	struct list_head ref_list;
+	struct rb_root ref_tree;
 	/* accumulate add BTRFS_ADD_DELAYED_REF nodes to this ref_add_list. */
 	struct list_head ref_add_list;
 
@@ -116,6 +107,14 @@ struct btrfs_delayed_ref_head {
 	int total_ref_mod;
 
 	/*
+	 * This is the current outstanding mod references for this bytenr.  This
+	 * is used with lookup_extent_info to get an accurate reference count
+	 * for a bytenr, so it is adjusted as delayed refs are run so that any
+	 * on disk reference count + ref_mod is accurate.
+	 */
+	int ref_mod;
+
+	/*
 	 * For qgroup reserved space freeing.
 	 *
 	 * ref_root and reserved will be recorded after
@@ -234,15 +233,18 @@ static inline void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref)
 		case BTRFS_SHARED_DATA_REF_KEY:
 			kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
 			break;
-		case 0:
-			kmem_cache_free(btrfs_delayed_ref_head_cachep, ref);
-			break;
 		default:
 			BUG();
 		}
 	}
 }
 
+static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *head)
+{
+	if (refcount_dec_and_test(&head->refs))
+		kmem_cache_free(btrfs_delayed_ref_head_cachep, head);
+}
+
 int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 			       struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes, u64 parent,
@@ -283,35 +285,17 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
 			    u64 seq);
 
 /*
- * a node might live in a head or a regular ref, this lets you
- * test for the proper type to use.
- */
-static int btrfs_delayed_ref_is_head(struct btrfs_delayed_ref_node *node)
-{
-	return node->is_head;
-}
-
-/*
  * helper functions to cast a node into its container
  */
 static inline struct btrfs_delayed_tree_ref *
 btrfs_delayed_node_to_tree_ref(struct btrfs_delayed_ref_node *node)
 {
-	WARN_ON(btrfs_delayed_ref_is_head(node));
 	return container_of(node, struct btrfs_delayed_tree_ref, node);
 }
 
 static inline struct btrfs_delayed_data_ref *
 btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node)
 {
-	WARN_ON(btrfs_delayed_ref_is_head(node));
 	return container_of(node, struct btrfs_delayed_data_ref, node);
 }
-
-static inline struct btrfs_delayed_ref_head *
-btrfs_delayed_node_to_head(struct btrfs_delayed_ref_node *node)
-{
-	WARN_ON(!btrfs_delayed_ref_is_head(node));
-	return container_of(node, struct btrfs_delayed_ref_head, node);
-}
 #endif