Merge tag 'ext4_for_linue' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 fixes from Ted Ts'o:
 "Fix a number of regression and other bugs in ext4, most of which were
  relatively obscure cornercases or races that were found using
  regression tests."

* tag 'ext4_for_linue' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (21 commits)
  ext4: fix data=journal fast mount/umount hang
  ext4: fix ext4_evict_inode() racing against workqueue processing code
  ext4: fix memory leakage in mext_check_coverage
  ext4: use s_extent_max_zeroout_kb value as number of kb
  ext4: use atomic64_t for the per-flexbg free_clusters count
  jbd2: fix use after free in jbd2_journal_dirty_metadata()
  ext4: reserve metadata block for every delayed write
  ext4: update reserved space after the 'correction'
  ext4: do not use yield()
  ext4: remove unused variable in ext4_free_blocks()
  ext4: fix WARN_ON from ext4_releasepage()
  ext4: fix the wrong number of the allocated blocks in ext4_split_extent()
  ext4: update extent status tree after an extent is zeroed out
  ext4: fix wrong m_len value after unwritten extent conversion
  ext4: add self-testing infrastructure to do a sanity check
  ext4: avoid a potential overflow in ext4_es_can_be_merged()
  ext4: invalidate extent status tree during extent migration
  ext4: remove unnecessary wait for extent conversion in ext4_fallocate()
  ext4: add warning to ext4_convert_unwritten_extents_endio
  ext4: disable merging of uninitialized extents
  ...

fs/ext4/ext4.h

@@ -335,9 +335,9 @@ struct ext4_group_desc
  */
 
 struct flex_groups {
-	atomic_t free_inodes;
-	atomic_t free_clusters;
-	atomic_t used_dirs;
+	atomic64_t	free_clusters;
+	atomic_t	free_inodes;
+	atomic_t	used_dirs;
 };
 
 #define EXT4_BG_INODE_UNINIT	0x0001 /* Inode table/bitmap not in use */
@@ -2617,7 +2617,7 @@ extern int ext4_move_extents(struct file *o_filp, struct file *d_filp,
 extern int __init ext4_init_pageio(void);
 extern void ext4_add_complete_io(ext4_io_end_t *io_end);
 extern void ext4_exit_pageio(void);
-extern void ext4_ioend_wait(struct inode *);
+extern void ext4_ioend_shutdown(struct inode *);
 extern void ext4_free_io_end(ext4_io_end_t *io);
 extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags);
 extern void ext4_end_io_work(struct work_struct *work);

fs/ext4/extents.c

@@ -1584,10 +1584,12 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
 	unsigned short ext1_ee_len, ext2_ee_len, max_len;
 
 	/*
-	 * Make sure that either both extents are uninitialized, or
-	 * both are _not_.
+	 * Make sure that both extents are initialized. We don't merge
+	 * uninitialized extents so that we can be sure that end_io code has
+	 * the extent that was written properly split out and conversion to
+	 * initialized is trivial.
 	 */
-	if (ext4_ext_is_uninitialized(ex1) ^ ext4_ext_is_uninitialized(ex2))
+	if (ext4_ext_is_uninitialized(ex1) || ext4_ext_is_uninitialized(ex2))
 		return 0;
 
 	if (ext4_ext_is_uninitialized(ex1))
@@ -2923,7 +2925,7 @@ static int ext4_split_extent_at(handle_t *handle,
 {
 	ext4_fsblk_t newblock;
 	ext4_lblk_t ee_block;
-	struct ext4_extent *ex, newex, orig_ex;
+	struct ext4_extent *ex, newex, orig_ex, zero_ex;
 	struct ext4_extent *ex2 = NULL;
 	unsigned int ee_len, depth;
 	int err = 0;
@@ -2943,6 +2945,10 @@ static int ext4_split_extent_at(handle_t *handle,
 	newblock = split - ee_block + ext4_ext_pblock(ex);
 
 	BUG_ON(split < ee_block || split >= (ee_block + ee_len));
+	BUG_ON(!ext4_ext_is_uninitialized(ex) &&
+	       split_flag & (EXT4_EXT_MAY_ZEROOUT |
+			     EXT4_EXT_MARK_UNINIT1 |
+			     EXT4_EXT_MARK_UNINIT2));
 
 	err = ext4_ext_get_access(handle, inode, path + depth);
 	if (err)
@@ -2990,12 +2996,26 @@ static int ext4_split_extent_at(handle_t *handle,
 	err = ext4_ext_insert_extent(handle, inode, path, &newex, flags);
 	if (err == -ENOSPC && (EXT4_EXT_MAY_ZEROOUT & split_flag)) {
 		if (split_flag & (EXT4_EXT_DATA_VALID1|EXT4_EXT_DATA_VALID2)) {
-			if (split_flag & EXT4_EXT_DATA_VALID1)
+			if (split_flag & EXT4_EXT_DATA_VALID1) {
 				err = ext4_ext_zeroout(inode, ex2);
-			else
+				zero_ex.ee_block = ex2->ee_block;
+				zero_ex.ee_len = ext4_ext_get_actual_len(ex2);
+				ext4_ext_store_pblock(&zero_ex,
+						      ext4_ext_pblock(ex2));
+			} else {
 				err = ext4_ext_zeroout(inode, ex);
-		} else
+				zero_ex.ee_block = ex->ee_block;
+				zero_ex.ee_len = ext4_ext_get_actual_len(ex);
+				ext4_ext_store_pblock(&zero_ex,
+						      ext4_ext_pblock(ex));
+			}
+		} else {
 			err = ext4_ext_zeroout(inode, &orig_ex);
+			zero_ex.ee_block = orig_ex.ee_block;
+			zero_ex.ee_len = ext4_ext_get_actual_len(&orig_ex);
+			ext4_ext_store_pblock(&zero_ex,
+					      ext4_ext_pblock(&orig_ex));
+		}
 
 		if (err)
 			goto fix_extent_len;
@@ -3003,6 +3023,12 @@ static int ext4_split_extent_at(handle_t *handle,
 		ex->ee_len = cpu_to_le16(ee_len);
 		ext4_ext_try_to_merge(handle, inode, path, ex);
 		err = ext4_ext_dirty(handle, inode, path + path->p_depth);
+		if (err)
+			goto fix_extent_len;
+
+		/* update extent status tree */
+		err = ext4_es_zeroout(inode, &zero_ex);
+
 		goto out;
 	} else if (err)
 		goto fix_extent_len;
@@ -3041,6 +3067,7 @@ static int ext4_split_extent(handle_t *handle,
 	int err = 0;
 	int uninitialized;
 	int split_flag1, flags1;
+	int allocated = map->m_len;
 
 	depth = ext_depth(inode);
 	ex = path[depth].p_ext;
@@ -3060,20 +3087,29 @@ static int ext4_split_extent(handle_t *handle,
 				map->m_lblk + map->m_len, split_flag1, flags1);
 		if (err)
 			goto out;
+	} else {
+		allocated = ee_len - (map->m_lblk - ee_block);
 	}
-
+	/*
+	 * Update path is required because previous ext4_split_extent_at() may
+	 * result in split of original leaf or extent zeroout.
+	 */
 	ext4_ext_drop_refs(path);
 	path = ext4_ext_find_extent(inode, map->m_lblk, path);
 	if (IS_ERR(path))
 		return PTR_ERR(path);
+	depth = ext_depth(inode);
+	ex = path[depth].p_ext;
+	uninitialized = ext4_ext_is_uninitialized(ex);
+	split_flag1 = 0;
 
 	if (map->m_lblk >= ee_block) {
-		split_flag1 = split_flag & (EXT4_EXT_MAY_ZEROOUT |
-					    EXT4_EXT_DATA_VALID2);
-		if (uninitialized)
+		split_flag1 = split_flag & EXT4_EXT_DATA_VALID2;
+		if (uninitialized) {
 			split_flag1 |= EXT4_EXT_MARK_UNINIT1;
-		if (split_flag & EXT4_EXT_MARK_UNINIT2)
-			split_flag1 |= EXT4_EXT_MARK_UNINIT2;
+			split_flag1 |= split_flag & (EXT4_EXT_MAY_ZEROOUT |
+						     EXT4_EXT_MARK_UNINIT2);
+		}
 		err = ext4_split_extent_at(handle, inode, path,
 				map->m_lblk, split_flag1, flags);
 		if (err)
@@ -3082,7 +3118,7 @@ static int ext4_split_extent(handle_t *handle,
 
 	ext4_ext_show_leaf(inode, path);
 out:
-	return err ? err : map->m_len;
+	return err ? err : allocated;
 }
 
 /*
@@ -3137,6 +3173,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 	ee_block = le32_to_cpu(ex->ee_block);
 	ee_len = ext4_ext_get_actual_len(ex);
 	allocated = ee_len - (map->m_lblk - ee_block);
+	zero_ex.ee_len = 0;
 
 	trace_ext4_ext_convert_to_initialized_enter(inode, map, ex);
 
@@ -3227,13 +3264,16 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 
 	if (EXT4_EXT_MAY_ZEROOUT & split_flag)
 		max_zeroout = sbi->s_extent_max_zeroout_kb >>
-			inode->i_sb->s_blocksize_bits;
+			(inode->i_sb->s_blocksize_bits - 10);
 
 	/* If extent is less than s_max_zeroout_kb, zeroout directly */
 	if (max_zeroout && (ee_len <= max_zeroout)) {
 		err = ext4_ext_zeroout(inode, ex);
 		if (err)
 			goto out;
+		zero_ex.ee_block = ex->ee_block;
+		zero_ex.ee_len = ext4_ext_get_actual_len(ex);
+		ext4_ext_store_pblock(&zero_ex, ext4_ext_pblock(ex));
 
 		err = ext4_ext_get_access(handle, inode, path + depth);
 		if (err)
@@ -3292,6 +3332,9 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 		err = allocated;
 
 out:
+	/* If we have gotten a failure, don't zero out status tree */
+	if (!err)
+		err = ext4_es_zeroout(inode, &zero_ex);
 	return err ? err : allocated;
 }
 
@@ -3374,8 +3417,19 @@ static int ext4_convert_unwritten_extents_endio(handle_t *handle,
 		"block %llu, max_blocks %u\n", inode->i_ino,
 		  (unsigned long long)ee_block, ee_len);
 
-	/* If extent is larger than requested then split is required */
+	/* If extent is larger than requested it is a clear sign that we still
+	 * have some extent state machine issues left. So extent_split is still
+	 * required.
+	 * TODO: Once all related issues will be fixed this situation should be
+	 * illegal.
+	 */
 	if (ee_block != map->m_lblk || ee_len > map->m_len) {
+#ifdef EXT4_DEBUG
+		ext4_warning("Inode (%ld) finished: extent logical block %llu,"
+			     " len %u; IO logical block %llu, len %u\n",
+			     inode->i_ino, (unsigned long long)ee_block, ee_len,
+			     (unsigned long long)map->m_lblk, map->m_len);
+#endif
 		err = ext4_split_unwritten_extents(handle, inode, map, path,
 						   EXT4_GET_BLOCKS_CONVERT);
 		if (err < 0)
@@ -3626,6 +3680,10 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 						 path, map->m_len);
 		} else
 			err = ret;
+		map->m_flags |= EXT4_MAP_MAPPED;
+		if (allocated > map->m_len)
+			allocated = map->m_len;
+		map->m_len = allocated;
 		goto out2;
 	}
 	/* buffered IO case */
@@ -3675,6 +3733,7 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 					allocated - map->m_len);
 		allocated = map->m_len;
 	}
+	map->m_len = allocated;
 
 	/*
 	 * If we have done fallocate with the offset that is already
@@ -4106,9 +4165,6 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			}
 		} else {
 			BUG_ON(allocated_clusters < reserved_clusters);
-			/* We will claim quota for all newly allocated blocks.*/
-			ext4_da_update_reserve_space(inode, allocated_clusters,
-							1);
 			if (reserved_clusters < allocated_clusters) {
 				struct ext4_inode_info *ei = EXT4_I(inode);
 				int reservation = allocated_clusters -
@@ -4159,6 +4215,15 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 				ei->i_reserved_data_blocks += reservation;
 				spin_unlock(&ei->i_block_reservation_lock);
 			}
+			/*
+			 * We will claim quota for all newly allocated blocks.
+			 * We're updating the reserved space *after* the
+			 * correction above so we do not accidentally free
+			 * all the metadata reservation because we might
+			 * actually need it later on.
+			 */
+			ext4_da_update_reserve_space(inode, allocated_clusters,
+							1);
 		}
 	}
 
@@ -4368,8 +4433,6 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 	if (len <= EXT_UNINIT_MAX_LEN << blkbits)
 		flags |= EXT4_GET_BLOCKS_NO_NORMALIZE;
 
-	/* Prevent race condition between unwritten */
-	ext4_flush_unwritten_io(inode);
 retry:
 	while (ret >= 0 && ret < max_blocks) {
 		map.m_lblk = map.m_lblk + ret;

fs/ext4/extents_status.c

@@ -333,17 +333,27 @@ static void ext4_es_free_extent(struct inode *inode, struct extent_status *es)
 static int ext4_es_can_be_merged(struct extent_status *es1,
 				 struct extent_status *es2)
 {
-	if (es1->es_lblk + es1->es_len != es2->es_lblk)
+	if (ext4_es_status(es1) != ext4_es_status(es2))
 		return 0;
 
-	if (ext4_es_status(es1) != ext4_es_status(es2))
+	if (((__u64) es1->es_len) + es2->es_len > 0xFFFFFFFFULL)
 		return 0;
 
-	if ((ext4_es_is_written(es1) || ext4_es_is_unwritten(es1)) &&
-	    (ext4_es_pblock(es1) + es1->es_len != ext4_es_pblock(es2)))
+	if (((__u64) es1->es_lblk) + es1->es_len != es2->es_lblk)
 		return 0;
 
-	return 1;
+	if ((ext4_es_is_written(es1) || ext4_es_is_unwritten(es1)) &&
+	    (ext4_es_pblock(es1) + es1->es_len == ext4_es_pblock(es2)))
+		return 1;
+
+	if (ext4_es_is_hole(es1))
+		return 1;
+
+	/* we need to check delayed extent is without unwritten status */
+	if (ext4_es_is_delayed(es1) && !ext4_es_is_unwritten(es1))
+		return 1;
+
+	return 0;
 }
 
 static struct extent_status *
@@ -389,6 +399,179 @@ ext4_es_try_to_merge_right(struct inode *inode, struct extent_status *es)
 	return es;
 }
 
+#ifdef ES_AGGRESSIVE_TEST
+static void ext4_es_insert_extent_ext_check(struct inode *inode,
+					    struct extent_status *es)
+{
+	struct ext4_ext_path *path = NULL;
+	struct ext4_extent *ex;
+	ext4_lblk_t ee_block;
+	ext4_fsblk_t ee_start;
+	unsigned short ee_len;
+	int depth, ee_status, es_status;
+
+	path = ext4_ext_find_extent(inode, es->es_lblk, NULL);
+	if (IS_ERR(path))
+		return;
+
+	depth = ext_depth(inode);
+	ex = path[depth].p_ext;
+
+	if (ex) {
+
+		ee_block = le32_to_cpu(ex->ee_block);
+		ee_start = ext4_ext_pblock(ex);
+		ee_len = ext4_ext_get_actual_len(ex);
+
+		ee_status = ext4_ext_is_uninitialized(ex) ? 1 : 0;
+		es_status = ext4_es_is_unwritten(es) ? 1 : 0;
+
+		/*
+		 * Make sure ex and es are not overlap when we try to insert
+		 * a delayed/hole extent.
+		 */
+		if (!ext4_es_is_written(es) && !ext4_es_is_unwritten(es)) {
+			if (in_range(es->es_lblk, ee_block, ee_len)) {
+				pr_warn("ES insert assertation failed for "
+					"inode: %lu we can find an extent "
+					"at block [%d/%d/%llu/%c], but we "
+					"want to add an delayed/hole extent "
+					"[%d/%d/%llu/%llx]\n",
+					inode->i_ino, ee_block, ee_len,
+					ee_start, ee_status ? 'u' : 'w',
+					es->es_lblk, es->es_len,
+					ext4_es_pblock(es), ext4_es_status(es));
+			}
+			goto out;
+		}
+
+		/*
+		 * We don't check ee_block == es->es_lblk, etc. because es
+		 * might be a part of whole extent, vice versa.
+		 */
+		if (es->es_lblk < ee_block ||
+		    ext4_es_pblock(es) != ee_start + es->es_lblk - ee_block) {
+			pr_warn("ES insert assertation failed for inode: %lu "
+				"ex_status [%d/%d/%llu/%c] != "
+				"es_status [%d/%d/%llu/%c]\n", inode->i_ino,
+				ee_block, ee_len, ee_start,
+				ee_status ? 'u' : 'w', es->es_lblk, es->es_len,
+				ext4_es_pblock(es), es_status ? 'u' : 'w');
+			goto out;
+		}
+
+		if (ee_status ^ es_status) {
+			pr_warn("ES insert assertation failed for inode: %lu "
+				"ex_status [%d/%d/%llu/%c] != "
+				"es_status [%d/%d/%llu/%c]\n", inode->i_ino,
+				ee_block, ee_len, ee_start,
+				ee_status ? 'u' : 'w', es->es_lblk, es->es_len,
+				ext4_es_pblock(es), es_status ? 'u' : 'w');
+		}
+	} else {
+		/*
+		 * We can't find an extent on disk.  So we need to make sure
+		 * that we don't want to add an written/unwritten extent.
+		 */
+		if (!ext4_es_is_delayed(es) && !ext4_es_is_hole(es)) {
+			pr_warn("ES insert assertation failed for inode: %lu "
+				"can't find an extent at block %d but we want "
+				"to add an written/unwritten extent "
+				"[%d/%d/%llu/%llx]\n", inode->i_ino,
+				es->es_lblk, es->es_lblk, es->es_len,
+				ext4_es_pblock(es), ext4_es_status(es));
+		}
+	}
+out:
+	if (path) {
+		ext4_ext_drop_refs(path);
+		kfree(path);
+	}
+}
+
+static void ext4_es_insert_extent_ind_check(struct inode *inode,
+					    struct extent_status *es)
+{
+	struct ext4_map_blocks map;
+	int retval;
+
+	/*
+	 * Here we call ext4_ind_map_blocks to lookup a block mapping because
+	 * 'Indirect' structure is defined in indirect.c.  So we couldn't
+	 * access direct/indirect tree from outside.  It is too dirty to define
+	 * this function in indirect.c file.
+	 */
+
+	map.m_lblk = es->es_lblk;
+	map.m_len = es->es_len;
+
+	retval = ext4_ind_map_blocks(NULL, inode, &map, 0);
+	if (retval > 0) {
+		if (ext4_es_is_delayed(es) || ext4_es_is_hole(es)) {
+			/*
+			 * We want to add a delayed/hole extent but this
+			 * block has been allocated.
+			 */
+			pr_warn("ES insert assertation failed for inode: %lu "
+				"We can find blocks but we want to add a "
+				"delayed/hole extent [%d/%d/%llu/%llx]\n",
+				inode->i_ino, es->es_lblk, es->es_len,
+				ext4_es_pblock(es), ext4_es_status(es));
+			return;
+		} else if (ext4_es_is_written(es)) {
+			if (retval != es->es_len) {
+				pr_warn("ES insert assertation failed for "
+					"inode: %lu retval %d != es_len %d\n",
+					inode->i_ino, retval, es->es_len);
+				return;
+			}
+			if (map.m_pblk != ext4_es_pblock(es)) {
+				pr_warn("ES insert assertation failed for "
+					"inode: %lu m_pblk %llu != "
+					"es_pblk %llu\n",
+					inode->i_ino, map.m_pblk,
+					ext4_es_pblock(es));
+				return;
+			}
+		} else {
+			/*
+			 * We don't need to check unwritten extent because
+			 * indirect-based file doesn't have it.
+			 */
+			BUG_ON(1);
+		}
+	} else if (retval == 0) {
+		if (ext4_es_is_written(es)) {
+			pr_warn("ES insert assertation failed for inode: %lu "
+				"We can't find the block but we want to add "
+				"an written extent [%d/%d/%llu/%llx]\n",
+				inode->i_ino, es->es_lblk, es->es_len,
+				ext4_es_pblock(es), ext4_es_status(es));
+			return;
+		}
+	}
+}
+
+static inline void ext4_es_insert_extent_check(struct inode *inode,
+					       struct extent_status *es)
+{
+	/*
+	 * We don't need to worry about the race condition because
+	 * caller takes i_data_sem locking.
+	 */
+	BUG_ON(!rwsem_is_locked(&EXT4_I(inode)->i_data_sem));
+	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
+		ext4_es_insert_extent_ext_check(inode, es);
+	else
+		ext4_es_insert_extent_ind_check(inode, es);
+}
+#else
+static inline void ext4_es_insert_extent_check(struct inode *inode,
+					       struct extent_status *es)
+{
+}
+#endif
+
 static int __es_insert_extent(struct inode *inode, struct extent_status *newes)
 {
 	struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree;
@@ -471,6 +654,8 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
 	ext4_es_store_status(&newes, status);
 	trace_ext4_es_insert_extent(inode, &newes);
 
+	ext4_es_insert_extent_check(inode, &newes);
+
 	write_lock(&EXT4_I(inode)->i_es_lock);
 	err = __es_remove_extent(inode, lblk, end);
 	if (err != 0)
@@ -669,6 +854,23 @@ int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 	return err;
 }
 
+int ext4_es_zeroout(struct inode *inode, struct ext4_extent *ex)
+{
+	ext4_lblk_t  ee_block;
+	ext4_fsblk_t ee_pblock;
+	unsigned int ee_len;
+
+	ee_block  = le32_to_cpu(ex->ee_block);
+	ee_len    = ext4_ext_get_actual_len(ex);
+	ee_pblock = ext4_ext_pblock(ex);
+
+	if (ee_len == 0)
+		return 0;
+
+	return ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock,
+				     EXTENT_STATUS_WRITTEN);
+}
+
 static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc)
 {
 	struct ext4_sb_info *sbi = container_of(shrink,

fs/ext4/extents_status.h

@@ -20,6 +20,12 @@
 #define es_debug(fmt, ...)	no_printk(fmt, ##__VA_ARGS__)
 #endif
 
+/*
+ * With ES_AGGRESSIVE_TEST defined, the result of es caching will be
+ * checked with old map_block's result.
+ */
+#define ES_AGGRESSIVE_TEST__
+
 /*
  * These flags live in the high bits of extent_status.es_pblk
  */
@@ -33,6 +39,8 @@
 				 EXTENT_STATUS_DELAYED | \
 				 EXTENT_STATUS_HOLE)
 
+struct ext4_extent;
+
 struct extent_status {
 	struct rb_node rb_node;
 	ext4_lblk_t es_lblk;	/* first logical block extent covers */
@@ -58,6 +66,7 @@ extern void ext4_es_find_delayed_extent(struct inode *inode, ext4_lblk_t lblk,
 					struct extent_status *es);
 extern int ext4_es_lookup_extent(struct inode *inode, ext4_lblk_t lblk,
 				 struct extent_status *es);
+extern int ext4_es_zeroout(struct inode *inode, struct ext4_extent *ex);
 
 static inline int ext4_es_is_written(struct extent_status *es)
 {

fs/ext4/ialloc.c

@@ -324,8 +324,8 @@ void ext4_free_inode(handle_t *handle, struct inode *inode)
 }
 
 struct orlov_stats {
+	__u64 free_clusters;
 	__u32 free_inodes;
-	__u32 free_clusters;
 	__u32 used_dirs;
 };
 
@@ -342,7 +342,7 @@ static void get_orlov_stats(struct super_block *sb, ext4_group_t g,
 
 	if (flex_size > 1) {
 		stats->free_inodes = atomic_read(&flex_group[g].free_inodes);
-		stats->free_clusters = atomic_read(&flex_group[g].free_clusters);
+		stats->free_clusters = atomic64_read(&flex_group[g].free_clusters);
 		stats->used_dirs = atomic_read(&flex_group[g].used_dirs);
 		return;
 	}

fs/ext4/inode.c

@@ -185,8 +185,6 @@ void ext4_evict_inode(struct inode *inode)
 
 	trace_ext4_evict_inode(inode);
 
-	ext4_ioend_wait(inode);
-
 	if (inode->i_nlink) {
 		/*
 		 * When journalling data dirty buffers are tracked only in the
@@ -207,7 +205,8 @@ void ext4_evict_inode(struct inode *inode)
 		 * don't use page cache.
 		 */
 		if (ext4_should_journal_data(inode) &&
-		    (S_ISLNK(inode->i_mode) || S_ISREG(inode->i_mode))) {
+		    (S_ISLNK(inode->i_mode) || S_ISREG(inode->i_mode)) &&
+		    inode->i_ino != EXT4_JOURNAL_INO) {
 			journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
 			tid_t commit_tid = EXT4_I(inode)->i_datasync_tid;
 
@@ -216,6 +215,7 @@ void ext4_evict_inode(struct inode *inode)
 			filemap_write_and_wait(&inode->i_data);
 		}
 		truncate_inode_pages(&inode->i_data, 0);
+		ext4_ioend_shutdown(inode);
 		goto no_delete;
 	}
 
@@ -225,6 +225,7 @@ void ext4_evict_inode(struct inode *inode)
 	if (ext4_should_order_data(inode))
 		ext4_begin_ordered_truncate(inode, 0);
 	truncate_inode_pages(&inode->i_data, 0);
+	ext4_ioend_shutdown(inode);
 
 	if (is_bad_inode(inode))
 		goto no_delete;
@@ -482,6 +483,58 @@ static pgoff_t ext4_num_dirty_pages(struct inode *inode, pgoff_t idx,
 	return num;
 }
 
+#ifdef ES_AGGRESSIVE_TEST
+static void ext4_map_blocks_es_recheck(handle_t *handle,
+				       struct inode *inode,
+				       struct ext4_map_blocks *es_map,
+				       struct ext4_map_blocks *map,
+				       int flags)
+{
+	int retval;
+
+	map->m_flags = 0;
+	/*
+	 * There is a race window that the result is not the same.
+	 * e.g. xfstests #223 when dioread_nolock enables.  The reason
+	 * is that we lookup a block mapping in extent status tree with
+	 * out taking i_data_sem.  So at the time the unwritten extent
+	 * could be converted.
+	 */
+	if (!(flags & EXT4_GET_BLOCKS_NO_LOCK))
+		down_read((&EXT4_I(inode)->i_data_sem));
+	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
+		retval = ext4_ext_map_blocks(handle, inode, map, flags &
+					     EXT4_GET_BLOCKS_KEEP_SIZE);
+	} else {
+		retval = ext4_ind_map_blocks(handle, inode, map, flags &
+					     EXT4_GET_BLOCKS_KEEP_SIZE);
+	}
+	if (!(flags & EXT4_GET_BLOCKS_NO_LOCK))
+		up_read((&EXT4_I(inode)->i_data_sem));
+	/*
+	 * Clear EXT4_MAP_FROM_CLUSTER and EXT4_MAP_BOUNDARY flag
+	 * because it shouldn't be marked in es_map->m_flags.
+	 */
+	map->m_flags &= ~(EXT4_MAP_FROM_CLUSTER | EXT4_MAP_BOUNDARY);
+
+	/*
+	 * We don't check m_len because extent will be collpased in status
+	 * tree.  So the m_len might not equal.
+	 */
+	if (es_map->m_lblk != map->m_lblk ||
+	    es_map->m_flags != map->m_flags ||
+	    es_map->m_pblk != map->m_pblk) {
+		printk("ES cache assertation failed for inode: %lu "
+		       "es_cached ex [%d/%d/%llu/%x] != "
+		       "found ex [%d/%d/%llu/%x] retval %d flags %x\n",
+		       inode->i_ino, es_map->m_lblk, es_map->m_len,
+		       es_map->m_pblk, es_map->m_flags, map->m_lblk,
+		       map->m_len, map->m_pblk, map->m_flags,
+		       retval, flags);
+	}
+}
+#endif /* ES_AGGRESSIVE_TEST */
+
 /*
  * The ext4_map_blocks() function tries to look up the requested blocks,
  * and returns if the blocks are already mapped.
@@ -509,6 +562,11 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 {
 	struct extent_status es;
 	int retval;
+#ifdef ES_AGGRESSIVE_TEST
+	struct ext4_map_blocks orig_map;
+
+	memcpy(&orig_map, map, sizeof(*map));
+#endif
 
 	map->m_flags = 0;
 	ext_debug("ext4_map_blocks(): inode %lu, flag %d, max_blocks %u,"
@@ -531,6 +589,10 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 		} else {
 			BUG_ON(1);
 		}
+#ifdef ES_AGGRESSIVE_TEST
+		ext4_map_blocks_es_recheck(handle, inode, map,
+					   &orig_map, flags);
+#endif
 		goto found;
 	}
 
@@ -551,6 +613,15 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 		int ret;
 		unsigned long long status;
 
+#ifdef ES_AGGRESSIVE_TEST
+		if (retval != map->m_len) {
+			printk("ES len assertation failed for inode: %lu "
+			       "retval %d != map->m_len %d "
+			       "in %s (lookup)\n", inode->i_ino, retval,
+			       map->m_len, __func__);
+		}
+#endif
+
 		status = map->m_flags & EXT4_MAP_UNWRITTEN ?
 				EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN;
 		if (!(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) &&
@@ -643,6 +714,24 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 		int ret;
 		unsigned long long status;
 
+#ifdef ES_AGGRESSIVE_TEST
+		if (retval != map->m_len) {
+			printk("ES len assertation failed for inode: %lu "
+			       "retval %d != map->m_len %d "
+			       "in %s (allocation)\n", inode->i_ino, retval,
+			       map->m_len, __func__);
+		}
+#endif
+
+		/*
+		 * If the extent has been zeroed out, we don't need to update
+		 * extent status tree.
+		 */
+		if ((flags & EXT4_GET_BLOCKS_PRE_IO) &&
+		    ext4_es_lookup_extent(inode, map->m_lblk, &es)) {
+			if (ext4_es_is_written(&es))
+				goto has_zeroout;
+		}
 		status = map->m_flags & EXT4_MAP_UNWRITTEN ?
 				EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN;
 		if (!(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) &&
@@ -655,6 +744,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 			retval = ret;
 	}
 
+has_zeroout:
 	up_write((&EXT4_I(inode)->i_data_sem));
 	if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) {
 		int ret = check_block_validity(inode, map);
@@ -1215,6 +1305,55 @@ static int ext4_journalled_write_end(struct file *file,
 	return ret ? ret : copied;
 }
 
+/*
+ * Reserve a metadata for a single block located at lblock
+ */
+static int ext4_da_reserve_metadata(struct inode *inode, ext4_lblk_t lblock)
+{
+	int retries = 0;
+	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	unsigned int md_needed;
+	ext4_lblk_t save_last_lblock;
+	int save_len;
+
+	/*
+	 * recalculate the amount of metadata blocks to reserve
+	 * in order to allocate nrblocks
+	 * worse case is one extent per block
+	 */
+repeat:
+	spin_lock(&ei->i_block_reservation_lock);
+	/*
+	 * ext4_calc_metadata_amount() has side effects, which we have
+	 * to be prepared undo if we fail to claim space.
+	 */
+	save_len = ei->i_da_metadata_calc_len;
+	save_last_lblock = ei->i_da_metadata_calc_last_lblock;
+	md_needed = EXT4_NUM_B2C(sbi,
+				 ext4_calc_metadata_amount(inode, lblock));
+	trace_ext4_da_reserve_space(inode, md_needed);
+
+	/*
+	 * We do still charge estimated metadata to the sb though;
+	 * we cannot afford to run out of free blocks.
+	 */
+	if (ext4_claim_free_clusters(sbi, md_needed, 0)) {
+		ei->i_da_metadata_calc_len = save_len;
+		ei->i_da_metadata_calc_last_lblock = save_last_lblock;
+		spin_unlock(&ei->i_block_reservation_lock);
+		if (ext4_should_retry_alloc(inode->i_sb, &retries)) {
+			cond_resched();
+			goto repeat;
+		}
+		return -ENOSPC;
+	}
+	ei->i_reserved_meta_blocks += md_needed;
+	spin_unlock(&ei->i_block_reservation_lock);
+
+	return 0;       /* success */
+}
+
 /*
  * Reserve a single cluster located at lblock
  */
@@ -1263,7 +1402,7 @@ static int ext4_da_reserve_space(struct inode *inode, ext4_lblk_t lblock)
 		ei->i_da_metadata_calc_last_lblock = save_last_lblock;
 		spin_unlock(&ei->i_block_reservation_lock);
 		if (ext4_should_retry_alloc(inode->i_sb, &retries)) {
-			yield();
+			cond_resched();
 			goto repeat;
 		}
 		dquot_release_reservation_block(inode, EXT4_C2B(sbi, 1));
@@ -1768,6 +1907,11 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock,
 	struct extent_status es;
 	int retval;
 	sector_t invalid_block = ~((sector_t) 0xffff);
+#ifdef ES_AGGRESSIVE_TEST
+	struct ext4_map_blocks orig_map;
+
+	memcpy(&orig_map, map, sizeof(*map));
+#endif
 
 	if (invalid_block < ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es))
 		invalid_block = ~0;
@@ -1809,6 +1953,9 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock,
 		else
 			BUG_ON(1);
 
+#ifdef ES_AGGRESSIVE_TEST
+		ext4_map_blocks_es_recheck(NULL, inode, map, &orig_map, 0);
+#endif
 		return retval;
 	}
 
@@ -1843,8 +1990,11 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock,
 		 * XXX: __block_prepare_write() unmaps passed block,
 		 * is it OK?
 		 */
-		/* If the block was allocated from previously allocated cluster,
-		 * then we dont need to reserve it again. */
+		/*
+		 * If the block was allocated from previously allocated cluster,
+		 * then we don't need to reserve it again. However we still need
+		 * to reserve metadata for every block we're going to write.
+		 */
 		if (!(map->m_flags & EXT4_MAP_FROM_CLUSTER)) {
 			ret = ext4_da_reserve_space(inode, iblock);
 			if (ret) {
@@ -1852,6 +2002,13 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock,
 				retval = ret;
 				goto out_unlock;
 			}
+		} else {
+			ret = ext4_da_reserve_metadata(inode, iblock);
+			if (ret) {
+				/* not enough space to reserve */
+				retval = ret;
+				goto out_unlock;
+			}
 		}
 
 		ret = ext4_es_insert_extent(inode, map->m_lblk, map->m_len,
@@ -1873,6 +2030,15 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock,
 		int ret;
 		unsigned long long status;
 
+#ifdef ES_AGGRESSIVE_TEST
+		if (retval != map->m_len) {
+			printk("ES len assertation failed for inode: %lu "
+			       "retval %d != map->m_len %d "
+			       "in %s (lookup)\n", inode->i_ino, retval,
+			       map->m_len, __func__);
+		}
+#endif
+
 		status = map->m_flags & EXT4_MAP_UNWRITTEN ?
 				EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN;
 		ret = ext4_es_insert_extent(inode, map->m_lblk, map->m_len,
@@ -2908,8 +3074,8 @@ static int ext4_releasepage(struct page *page, gfp_t wait)
 
 	trace_ext4_releasepage(page);
 
-	WARN_ON(PageChecked(page));
-	if (!page_has_buffers(page))
+	/* Page has dirty journalled data -> cannot release */
+	if (PageChecked(page))
 		return 0;
 	if (journal)
 		return jbd2_journal_try_to_free_buffers(journal, page, wait);

fs/ext4/mballoc.c

@@ -2804,8 +2804,8 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi,
 							  ac->ac_b_ex.fe_group);
-		atomic_sub(ac->ac_b_ex.fe_len,
-			   &sbi->s_flex_groups[flex_group].free_clusters);
+		atomic64_sub(ac->ac_b_ex.fe_len,
+			     &sbi->s_flex_groups[flex_group].free_clusters);
 	}
 
 	err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
@@ -3692,11 +3692,7 @@ ext4_mb_discard_group_preallocations(struct super_block *sb,
 	if (free < needed && busy) {
 		busy = 0;
 		ext4_unlock_group(sb, group);
-		/*
-		 * Yield the CPU here so that we don't get soft lockup
-		 * in non preempt case.
-		 */
-		yield();
+		cond_resched();
 		goto repeat;
 	}
 
@@ -4246,7 +4242,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 			ext4_claim_free_clusters(sbi, ar->len, ar->flags)) {
 
 			/* let others to free the space */
-			yield();
+			cond_resched();
 			ar->len = ar->len >> 1;
 		}
 		if (!ar->len) {
@@ -4464,7 +4460,6 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 	struct buffer_head *bitmap_bh = NULL;
 	struct super_block *sb = inode->i_sb;
 	struct ext4_group_desc *gdp;
-	unsigned long freed = 0;
 	unsigned int overflow;
 	ext4_grpblk_t bit;
 	struct buffer_head *gd_bh;
@@ -4666,14 +4661,12 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
-		atomic_add(count_clusters,
-			   &sbi->s_flex_groups[flex_group].free_clusters);
+		atomic64_add(count_clusters,
+			     &sbi->s_flex_groups[flex_group].free_clusters);
 	}
 
 	ext4_mb_unload_buddy(&e4b);
 
-	freed += count;
-
 	if (!(flags & EXT4_FREE_BLOCKS_NO_QUOT_UPDATE))
 		dquot_free_block(inode, EXT4_C2B(sbi, count_clusters));
 
@@ -4811,8 +4804,8 @@ int ext4_group_add_blocks(handle_t *handle, struct super_block *sb,
 
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
-		atomic_add(EXT4_NUM_B2C(sbi, blocks_freed),
-			   &sbi->s_flex_groups[flex_group].free_clusters);
+		atomic64_add(EXT4_NUM_B2C(sbi, blocks_freed),
+			     &sbi->s_flex_groups[flex_group].free_clusters);
 	}
 
 	ext4_mb_unload_buddy(&e4b);

fs/ext4/move_extent.c

@@ -32,16 +32,18 @@
  */
 static inline int
 get_ext_path(struct inode *inode, ext4_lblk_t lblock,
-		struct ext4_ext_path **path)
+		struct ext4_ext_path **orig_path)
 {
 	int ret = 0;
+	struct ext4_ext_path *path;
 
-	*path = ext4_ext_find_extent(inode, lblock, *path);
-	if (IS_ERR(*path)) {
-		ret = PTR_ERR(*path);
-		*path = NULL;
-	} else if ((*path)[ext_depth(inode)].p_ext == NULL)
+	path = ext4_ext_find_extent(inode, lblock, *orig_path);
+	if (IS_ERR(path))
+		ret = PTR_ERR(path);
+	else if (path[ext_depth(inode)].p_ext == NULL)
 		ret = -ENODATA;
+	else
+		*orig_path = path;
 
 	return ret;
 }
@@ -611,24 +613,25 @@ mext_check_coverage(struct inode *inode, ext4_lblk_t from, ext4_lblk_t count,
 {
 	struct ext4_ext_path *path = NULL;
 	struct ext4_extent *ext;
+	int ret = 0;
 	ext4_lblk_t last = from + count;
 	while (from < last) {
 		*err = get_ext_path(inode, from, &path);
 		if (*err)
-			return 0;
+			goto out;
 		ext = path[ext_depth(inode)].p_ext;
-		if (!ext) {
-			ext4_ext_drop_refs(path);
-			return 0;
-		}
-		if (uninit != ext4_ext_is_uninitialized(ext)) {
-			ext4_ext_drop_refs(path);
-			return 0;
-		}
+		if (uninit != ext4_ext_is_uninitialized(ext))
+			goto out;
 		from += ext4_ext_get_actual_len(ext);
 		ext4_ext_drop_refs(path);
 	}
-	return 1;
+	ret = 1;
+out:
+	if (path) {
+		ext4_ext_drop_refs(path);
+		kfree(path);
+	}
+	return ret;
 }
 
 /**
@@ -666,6 +669,14 @@ mext_replace_branches(handle_t *handle, struct inode *orig_inode,
 	int replaced_count = 0;
 	int dext_alen;
 
+	*err = ext4_es_remove_extent(orig_inode, from, count);
+	if (*err)
+		goto out;
+
+	*err = ext4_es_remove_extent(donor_inode, from, count);
+	if (*err)
+		goto out;
+
 	/* Get the original extent for the block "orig_off" */
 	*err = get_ext_path(orig_inode, orig_off, &orig_path);
 	if (*err)

fs/ext4/page-io.c

@@ -50,11 +50,21 @@ void ext4_exit_pageio(void)
 	kmem_cache_destroy(io_page_cachep);
 }
 
-void ext4_ioend_wait(struct inode *inode)
+/*
+ * This function is called by ext4_evict_inode() to make sure there is
+ * no more pending I/O completion work left to do.
+ */
+void ext4_ioend_shutdown(struct inode *inode)
 {
 	wait_queue_head_t *wq = ext4_ioend_wq(inode);
 
 	wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_ioend_count) == 0));
+	/*
+	 * We need to make sure the work structure is finished being
+	 * used before we let the inode get destroyed.
+	 */
+	if (work_pending(&EXT4_I(inode)->i_unwritten_work))
+		cancel_work_sync(&EXT4_I(inode)->i_unwritten_work);
 }
 
 static void put_io_page(struct ext4_io_page *io_page)

fs/ext4/resize.c

@@ -1360,8 +1360,8 @@ static void ext4_update_super(struct super_block *sb,
 	    sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group;
 		flex_group = ext4_flex_group(sbi, group_data[0].group);
-		atomic_add(EXT4_NUM_B2C(sbi, free_blocks),
-			   &sbi->s_flex_groups[flex_group].free_clusters);
+		atomic64_add(EXT4_NUM_B2C(sbi, free_blocks),
+			     &sbi->s_flex_groups[flex_group].free_clusters);
 		atomic_add(EXT4_INODES_PER_GROUP(sb) * flex_gd->count,
 			   &sbi->s_flex_groups[flex_group].free_inodes);
 	}

fs/ext4/super.c

@@ -1927,8 +1927,8 @@ static int ext4_fill_flex_info(struct super_block *sb)
 		flex_group = ext4_flex_group(sbi, i);
 		atomic_add(ext4_free_inodes_count(sb, gdp),
 			   &sbi->s_flex_groups[flex_group].free_inodes);
-		atomic_add(ext4_free_group_clusters(sb, gdp),
-			   &sbi->s_flex_groups[flex_group].free_clusters);
+		atomic64_add(ext4_free_group_clusters(sb, gdp),
+			     &sbi->s_flex_groups[flex_group].free_clusters);
 		atomic_add(ext4_used_dirs_count(sb, gdp),
 			   &sbi->s_flex_groups[flex_group].used_dirs);
 	}

fs/jbd2/transaction.c

@@ -1065,9 +1065,12 @@ int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
 void jbd2_journal_set_triggers(struct buffer_head *bh,
 			       struct jbd2_buffer_trigger_type *type)
 {
-	struct journal_head *jh = bh2jh(bh);
+	struct journal_head *jh = jbd2_journal_grab_journal_head(bh);
 
+	if (WARN_ON(!jh))
+		return;
 	jh->b_triggers = type;
+	jbd2_journal_put_journal_head(jh);
 }
 
 void jbd2_buffer_frozen_trigger(struct journal_head *jh, void *mapped_data,
@@ -1119,17 +1122,18 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
-	struct journal_head *jh = bh2jh(bh);
+	struct journal_head *jh;
 	int ret = 0;
 
-	jbd_debug(5, "journal_head %p\n", jh);
-	JBUFFER_TRACE(jh, "entry");
 	if (is_handle_aborted(handle))
 		goto out;
-	if (!buffer_jbd(bh)) {
+	jh = jbd2_journal_grab_journal_head(bh);
+	if (!jh) {
 		ret = -EUCLEAN;
 		goto out;
 	}
+	jbd_debug(5, "journal_head %p\n", jh);
+	JBUFFER_TRACE(jh, "entry");
 
 	jbd_lock_bh_state(bh);
 
@@ -1220,6 +1224,7 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 	spin_unlock(&journal->j_list_lock);
 out_unlock_bh:
 	jbd_unlock_bh_state(bh);
+	jbd2_journal_put_journal_head(jh);
 out:
 	JBUFFER_TRACE(jh, "exit");
 	WARN_ON(ret);	/* All errors are bugs, so dump the stack */