Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates from Ted Ts'o:
 "Add new ioctls to set and get the file system UUID in the ext4
  superblock and improved the performance of the online resizing of file
  systems with bigalloc enabled.

  Fixed a lot of bugs, in particular for the inline data feature,
  potential races when creating and deleting inodes with shared extended
  attribute blocks, and the handling of directory blocks which are
  corrupted"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (37 commits)
  ext4: add ioctls to get/set the ext4 superblock uuid
  ext4: avoid resizing to a partial cluster size
  ext4: reduce computation of overhead during resize
  jbd2: fix assertion 'jh->b_frozen_data == NULL' failure when journal aborted
  ext4: block range must be validated before use in ext4_mb_clear_bb()
  mbcache: automatically delete entries from cache on freeing
  mbcache: Remove mb_cache_entry_delete()
  ext2: avoid deleting xattr block that is being reused
  ext2: unindent codeblock in ext2_xattr_set()
  ext2: factor our freeing of xattr block reference
  ext4: fix race when reusing xattr blocks
  ext4: unindent codeblock in ext4_xattr_block_set()
  ext4: remove EA inode entry from mbcache on inode eviction
  mbcache: add functions to delete entry if unused
  mbcache: don't reclaim used entries
  ext4: make sure ext4_append() always allocates new block
  ext4: check if directory block is within i_size
  ext4: reflect mb_optimize_scan value in options file
  ext4: avoid remove directory when directory is corrupted
  ext4: aligned '*' in comments
  ...

1 files changed, 63 insertions, 60 deletions

diff --git a/fs/mbcache.c b/fs/mbcache.c
index 97c54d3a2227..96f1d49d30a5 100644
--- a/fs/mbcache.c
+++ b/fs/mbcache.c
@@ -11,7 +11,7 @@
 /*
  * Mbcache is a simple key-value store. Keys need not be unique, however
  * key-value pairs are expected to be unique (we use this fact in
- * mb_cache_entry_delete()).
+ * mb_cache_entry_delete_or_get()).
  *
  * Ext2 and ext4 use this cache for deduplication of extended attribute blocks.
  * Ext4 also uses it for deduplication of xattr values stored in inodes.
@@ -90,7 +90,7 @@ int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key,
 		return -ENOMEM;
 
 	INIT_LIST_HEAD(&entry->e_list);
-	/* One ref for hash, one ref returned */
+	/* Initial hash reference */
 	atomic_set(&entry->e_refcnt, 1);
 	entry->e_key = key;
 	entry->e_value = value;
@@ -106,25 +106,45 @@ int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key,
 		}
 	}
 	hlist_bl_add_head(&entry->e_hash_list, head);
-	hlist_bl_unlock(head);
-
+	/*
+	 * Add entry to LRU list before it can be found by
+	 * mb_cache_entry_delete() to avoid races
+	 */
 	spin_lock(&cache->c_list_lock);
 	list_add_tail(&entry->e_list, &cache->c_list);
-	/* Grab ref for LRU list */
-	atomic_inc(&entry->e_refcnt);
 	cache->c_entry_count++;
 	spin_unlock(&cache->c_list_lock);
+	hlist_bl_unlock(head);
 
 	return 0;
 }
 EXPORT_SYMBOL(mb_cache_entry_create);
 
-void __mb_cache_entry_free(struct mb_cache_entry *entry)
+void __mb_cache_entry_free(struct mb_cache *cache, struct mb_cache_entry *entry)
 {
+	struct hlist_bl_head *head;
+
+	head = mb_cache_entry_head(cache, entry->e_key);
+	hlist_bl_lock(head);
+	hlist_bl_del(&entry->e_hash_list);
+	hlist_bl_unlock(head);
 	kmem_cache_free(mb_entry_cache, entry);
 }
 EXPORT_SYMBOL(__mb_cache_entry_free);
 
+/*
+ * mb_cache_entry_wait_unused - wait to be the last user of the entry
+ *
+ * @entry - entry to work on
+ *
+ * Wait to be the last user of the entry.
+ */
+void mb_cache_entry_wait_unused(struct mb_cache_entry *entry)
+{
+	wait_var_event(&entry->e_refcnt, atomic_read(&entry->e_refcnt) <= 2);
+}
+EXPORT_SYMBOL(mb_cache_entry_wait_unused);
+
 static struct mb_cache_entry *__entry_find(struct mb_cache *cache,
 					   struct mb_cache_entry *entry,
 					   u32 key)
@@ -142,10 +162,9 @@ static struct mb_cache_entry *__entry_find(struct mb_cache *cache,
 	while (node) {
 		entry = hlist_bl_entry(node, struct mb_cache_entry,
 				       e_hash_list);
-		if (entry->e_key == key && entry->e_reusable) {
-			atomic_inc(&entry->e_refcnt);
+		if (entry->e_key == key && entry->e_reusable &&
+		    atomic_inc_not_zero(&entry->e_refcnt))
 			goto out;
-		}
 		node = node->next;
 	}
 	entry = NULL;
@@ -205,10 +224,9 @@ struct mb_cache_entry *mb_cache_entry_get(struct mb_cache *cache, u32 key,
 	head = mb_cache_entry_head(cache, key);
 	hlist_bl_lock(head);
 	hlist_bl_for_each_entry(entry, node, head, e_hash_list) {
-		if (entry->e_key == key && entry->e_value == value) {
-			atomic_inc(&entry->e_refcnt);
+		if (entry->e_key == key && entry->e_value == value &&
+		    atomic_inc_not_zero(&entry->e_refcnt))
 			goto out;
-		}
 	}
 	entry = NULL;
 out:
@@ -217,42 +235,42 @@ out:
 }
 EXPORT_SYMBOL(mb_cache_entry_get);
 
-/* mb_cache_entry_delete - remove a cache entry
+/* mb_cache_entry_delete_or_get - remove a cache entry if it has no users
  * @cache - cache we work with
  * @key - key
  * @value - value
  *
- * Remove entry from cache @cache with key @key and value @value.
+ * Remove entry from cache @cache with key @key and value @value. The removal
+ * happens only if the entry is unused. The function returns NULL in case the
+ * entry was successfully removed or there's no entry in cache. Otherwise the
+ * function grabs reference of the entry that we failed to delete because it
+ * still has users and return it.
  */
-void mb_cache_entry_delete(struct mb_cache *cache, u32 key, u64 value)
+struct mb_cache_entry *mb_cache_entry_delete_or_get(struct mb_cache *cache,
+						    u32 key, u64 value)
 {
-	struct hlist_bl_node *node;
-	struct hlist_bl_head *head;
 	struct mb_cache_entry *entry;
 
-	head = mb_cache_entry_head(cache, key);
-	hlist_bl_lock(head);
-	hlist_bl_for_each_entry(entry, node, head, e_hash_list) {
-		if (entry->e_key == key && entry->e_value == value) {
-			/* We keep hash list reference to keep entry alive */
-			hlist_bl_del_init(&entry->e_hash_list);
-			hlist_bl_unlock(head);
-			spin_lock(&cache->c_list_lock);
-			if (!list_empty(&entry->e_list)) {
-				list_del_init(&entry->e_list);
-				if (!WARN_ONCE(cache->c_entry_count == 0,
-		"mbcache: attempt to decrement c_entry_count past zero"))
-					cache->c_entry_count--;
-				atomic_dec(&entry->e_refcnt);
-			}
-			spin_unlock(&cache->c_list_lock);
-			mb_cache_entry_put(cache, entry);
-			return;
-		}
-	}
-	hlist_bl_unlock(head);
+	entry = mb_cache_entry_get(cache, key, value);
+	if (!entry)
+		return NULL;
+
+	/*
+	 * Drop the ref we got from mb_cache_entry_get() and the initial hash
+	 * ref if we are the last user
+	 */
+	if (atomic_cmpxchg(&entry->e_refcnt, 2, 0) != 2)
+		return entry;
+
+	spin_lock(&cache->c_list_lock);
+	if (!list_empty(&entry->e_list))
+		list_del_init(&entry->e_list);
+	cache->c_entry_count--;
+	spin_unlock(&cache->c_list_lock);
+	__mb_cache_entry_free(cache, entry);
+	return NULL;
 }
-EXPORT_SYMBOL(mb_cache_entry_delete);
+EXPORT_SYMBOL(mb_cache_entry_delete_or_get);
 
 /* mb_cache_entry_touch - cache entry got used
  * @cache - cache the entry belongs to
@@ -281,34 +299,24 @@ static unsigned long mb_cache_shrink(struct mb_cache *cache,
 				     unsigned long nr_to_scan)
 {
 	struct mb_cache_entry *entry;
-	struct hlist_bl_head *head;
 	unsigned long shrunk = 0;
 
 	spin_lock(&cache->c_list_lock);
 	while (nr_to_scan-- && !list_empty(&cache->c_list)) {
 		entry = list_first_entry(&cache->c_list,
 					 struct mb_cache_entry, e_list);
-		if (entry->e_referenced) {
+		/* Drop initial hash reference if there is no user */
+		if (entry->e_referenced ||
+		    atomic_cmpxchg(&entry->e_refcnt, 1, 0) != 1) {
 			entry->e_referenced = 0;
 			list_move_tail(&entry->e_list, &cache->c_list);
 			continue;
 		}
 		list_del_init(&entry->e_list);
 		cache->c_entry_count--;
-		/*
-		 * We keep LRU list reference so that entry doesn't go away
-		 * from under us.
-		 */
 		spin_unlock(&cache->c_list_lock);
-		head = mb_cache_entry_head(cache, entry->e_key);
-		hlist_bl_lock(head);
-		if (!hlist_bl_unhashed(&entry->e_hash_list)) {
-			hlist_bl_del_init(&entry->e_hash_list);
-			atomic_dec(&entry->e_refcnt);
-		}
-		hlist_bl_unlock(head);
-		if (mb_cache_entry_put(cache, entry))
-			shrunk++;
+		__mb_cache_entry_free(cache, entry);
+		shrunk++;
 		cond_resched();
 		spin_lock(&cache->c_list_lock);
 	}
@@ -400,11 +408,6 @@ void mb_cache_destroy(struct mb_cache *cache)
 	 * point.
 	 */
 	list_for_each_entry_safe(entry, next, &cache->c_list, e_list) {
-		if (!hlist_bl_unhashed(&entry->e_hash_list)) {
-			hlist_bl_del_init(&entry->e_hash_list);
-			atomic_dec(&entry->e_refcnt);
-		} else
-			WARN_ON(1);
 		list_del(&entry->e_list);
 		WARN_ON(atomic_read(&entry->e_refcnt) != 1);
 		mb_cache_entry_put(cache, entry);