Merge tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 fixes from Ted Ts'o:
 "More miscellaneous ext4 bug fixes (all stable fodder)"

* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix mount failure with quota configured as module
  jbd2: fix ocfs2 corrupt when clearing block group bits
  ext4: fix race between writepages and enabling EXT4_EXTENTS_FL
  ext4: rename s_journal_flag_rwsem to s_writepages_rwsem
  ext4: fix potential race between s_flex_groups online resizing and access
  ext4: fix potential race between s_group_info online resizing and access
  ext4: fix potential race between online resizing and write operations
  ext4: add cond_resched() to __ext4_find_entry()
  ext4: fix a data race in EXT4_I(inode)->i_disksize

Author: torvalds

fs/ext4/balloc.c

@@ -270,6 +270,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb,
 	ext4_group_t ngroups = ext4_get_groups_count(sb);
 	struct ext4_group_desc *desc;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct buffer_head *bh_p;
 
 	if (block_group >= ngroups) {
 		ext4_error(sb, "block_group >= groups_count - block_group = %u,"
@@ -280,18 +281,25 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb,
 
 	group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb);
 	offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1);
-	if (!sbi->s_group_desc[group_desc]) {
+	bh_p = sbi_array_rcu_deref(sbi, s_group_desc, group_desc);
+	/*
+	 * sbi_array_rcu_deref returns with rcu unlocked, this is ok since
+	 * the pointer being dereferenced won't be dereferenced again. By
+	 * looking at the usage in add_new_gdb() the value isn't modified,
+	 * just the pointer, and so it remains valid.
+	 */
+	if (!bh_p) {
 		ext4_error(sb, "Group descriptor not loaded - "
 			   "block_group = %u, group_desc = %u, desc = %u",
 			   block_group, group_desc, offset);
 		return NULL;
 	}
 
 	desc = (struct ext4_group_desc *)(
-		(__u8 *)sbi->s_group_desc[group_desc]->b_data +
+		(__u8 *)bh_p->b_data +
 		offset * EXT4_DESC_SIZE(sb));
 	if (bh)
-		*bh = sbi->s_group_desc[group_desc];
+		*bh = bh_p;
 	return desc;
 }
 

fs/ext4/ext4.h

@@ -1400,7 +1400,7 @@ struct ext4_sb_info {
 	loff_t s_bitmap_maxbytes;	/* max bytes for bitmap files */
 	struct buffer_head * s_sbh;	/* Buffer containing the super block */
 	struct ext4_super_block *s_es;	/* Pointer to the super block in the buffer */
-	struct buffer_head **s_group_desc;
+	struct buffer_head * __rcu *s_group_desc;
 	unsigned int s_mount_opt;
 	unsigned int s_mount_opt2;
 	unsigned int s_mount_flags;
@@ -1462,7 +1462,7 @@ struct ext4_sb_info {
 #endif
 
 	/* for buddy allocator */
-	struct ext4_group_info ***s_group_info;
+	struct ext4_group_info ** __rcu *s_group_info;
 	struct inode *s_buddy_cache;
 	spinlock_t s_md_lock;
 	unsigned short *s_mb_offsets;
@@ -1512,7 +1512,7 @@ struct ext4_sb_info {
 	unsigned int s_extent_max_zeroout_kb;
 
 	unsigned int s_log_groups_per_flex;
-	struct flex_groups *s_flex_groups;
+	struct flex_groups * __rcu *s_flex_groups;
 	ext4_group_t s_flex_groups_allocated;
 
 	/* workqueue for reserved extent conversions (buffered io) */
@@ -1552,8 +1552,11 @@ struct ext4_sb_info {
 	struct ratelimit_state s_warning_ratelimit_state;
 	struct ratelimit_state s_msg_ratelimit_state;
 
-	/* Barrier between changing inodes' journal flags and writepages ops. */
-	struct percpu_rw_semaphore s_journal_flag_rwsem;
+	/*
+	 * Barrier between writepages ops and changing any inode's JOURNAL_DATA
+	 * or EXTENTS flag.
+	 */
+	struct percpu_rw_semaphore s_writepages_rwsem;
 	struct dax_device *s_daxdev;
 #ifdef CONFIG_EXT4_DEBUG
 	unsigned long s_simulate_fail;
@@ -1576,6 +1579,23 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 		 ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count));
 }
 
+/*
+ * Returns: sbi->field[index]
+ * Used to access an array element from the following sbi fields which require
+ * rcu protection to avoid dereferencing an invalid pointer due to reassignment
+ * - s_group_desc
+ * - s_group_info
+ * - s_flex_group
+ */
+#define sbi_array_rcu_deref(sbi, field, index)				   \
+({									   \
+	typeof(*((sbi)->field)) _v;					   \
+	rcu_read_lock();						   \
+	_v = ((typeof(_v)*)rcu_dereference((sbi)->field))[index];	   \
+	rcu_read_unlock();						   \
+	_v;								   \
+})
+
 /*
  * Simulate_fail codes
  */
@@ -2730,6 +2750,7 @@ extern int ext4_generic_delete_entry(handle_t *handle,
 extern bool ext4_empty_dir(struct inode *inode);
 
 /* resize.c */
+extern void ext4_kvfree_array_rcu(void *to_free);
 extern int ext4_group_add(struct super_block *sb,
 				struct ext4_new_group_data *input);
 extern int ext4_group_extend(struct super_block *sb,
@@ -2976,13 +2997,13 @@ static inline
 struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
 					    ext4_group_t group)
 {
-	 struct ext4_group_info ***grp_info;
+	 struct ext4_group_info **grp_info;
 	 long indexv, indexh;
 	 BUG_ON(group >= EXT4_SB(sb)->s_groups_count);
-	 grp_info = EXT4_SB(sb)->s_group_info;
 	 indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb));
 	 indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1);
-	 return grp_info[indexv][indexh];
+	 grp_info = sbi_array_rcu_deref(EXT4_SB(sb), s_group_info, indexv);
+	 return grp_info[indexh];
 }
 
 /*
@@ -3032,7 +3053,7 @@ static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize)
 		     !inode_is_locked(inode));
 	down_write(&EXT4_I(inode)->i_data_sem);
 	if (newsize > EXT4_I(inode)->i_disksize)
-		EXT4_I(inode)->i_disksize = newsize;
+		WRITE_ONCE(EXT4_I(inode)->i_disksize, newsize);
 	up_write(&EXT4_I(inode)->i_data_sem);
 }
 

fs/ext4/ialloc.c

@@ -328,11 +328,13 @@ void ext4_free_inode(handle_t *handle, struct inode *inode)
 
 	percpu_counter_inc(&sbi->s_freeinodes_counter);
 	if (sbi->s_log_groups_per_flex) {
-		ext4_group_t f = ext4_flex_group(sbi, block_group);
+		struct flex_groups *fg;
 
-		atomic_inc(&sbi->s_flex_groups[f].free_inodes);
+		fg = sbi_array_rcu_deref(sbi, s_flex_groups,
+					 ext4_flex_group(sbi, block_group));
+		atomic_inc(&fg->free_inodes);
 		if (is_directory)
-			atomic_dec(&sbi->s_flex_groups[f].used_dirs);
+			atomic_dec(&fg->used_dirs);
 	}
 	BUFFER_TRACE(bh2, "call ext4_handle_dirty_metadata");
 	fatal = ext4_handle_dirty_metadata(handle, NULL, bh2);
@@ -368,12 +370,13 @@ static void get_orlov_stats(struct super_block *sb, ext4_group_t g,
 			    int flex_size, struct orlov_stats *stats)
 {
 	struct ext4_group_desc *desc;
-	struct flex_groups *flex_group = EXT4_SB(sb)->s_flex_groups;
 
 	if (flex_size > 1) {
-		stats->free_inodes = atomic_read(&flex_group[g].free_inodes);
-		stats->free_clusters = atomic64_read(&flex_group[g].free_clusters);
-		stats->used_dirs = atomic_read(&flex_group[g].used_dirs);
+		struct flex_groups *fg = sbi_array_rcu_deref(EXT4_SB(sb),
+							     s_flex_groups, g);
+		stats->free_inodes = atomic_read(&fg->free_inodes);
+		stats->free_clusters = atomic64_read(&fg->free_clusters);
+		stats->used_dirs = atomic_read(&fg->used_dirs);
 		return;
 	}
 
@@ -1054,7 +1057,8 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
 		if (sbi->s_log_groups_per_flex) {
 			ext4_group_t f = ext4_flex_group(sbi, group);
 
-			atomic_inc(&sbi->s_flex_groups[f].used_dirs);
+			atomic_inc(&sbi_array_rcu_deref(sbi, s_flex_groups,
+							f)->used_dirs);
 		}
 	}
 	if (ext4_has_group_desc_csum(sb)) {
@@ -1077,7 +1081,8 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
 
 	if (sbi->s_log_groups_per_flex) {
 		flex_group = ext4_flex_group(sbi, group);
-		atomic_dec(&sbi->s_flex_groups[flex_group].free_inodes);
+		atomic_dec(&sbi_array_rcu_deref(sbi, s_flex_groups,
+						flex_group)->free_inodes);
 	}
 
 	inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);

fs/ext4/inode.c

@@ -2465,7 +2465,7 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 	 * truncate are avoided by checking i_size under i_data_sem.
 	 */
 	disksize = ((loff_t)mpd->first_page) << PAGE_SHIFT;
-	if (disksize > EXT4_I(inode)->i_disksize) {
+	if (disksize > READ_ONCE(EXT4_I(inode)->i_disksize)) {
 		int err2;
 		loff_t i_size;
 
@@ -2628,7 +2628,7 @@ static int ext4_writepages(struct address_space *mapping,
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return -EIO;
 
-	percpu_down_read(&sbi->s_journal_flag_rwsem);
+	percpu_down_read(&sbi->s_writepages_rwsem);
 	trace_ext4_writepages(inode, wbc);
 
 	/*
@@ -2849,7 +2849,7 @@ static int ext4_writepages(struct address_space *mapping,
 out_writepages:
 	trace_ext4_writepages_result(inode, wbc, ret,
 				     nr_to_write - wbc->nr_to_write);
-	percpu_up_read(&sbi->s_journal_flag_rwsem);
+	percpu_up_read(&sbi->s_writepages_rwsem);
 	return ret;
 }
 
@@ -2864,13 +2864,13 @@ static int ext4_dax_writepages(struct address_space *mapping,
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return -EIO;
 
-	percpu_down_read(&sbi->s_journal_flag_rwsem);
+	percpu_down_read(&sbi->s_writepages_rwsem);
 	trace_ext4_writepages(inode, wbc);
 
 	ret = dax_writeback_mapping_range(mapping, sbi->s_daxdev, wbc);
 	trace_ext4_writepages_result(inode, wbc, ret,
 				     nr_to_write - wbc->nr_to_write);
-	percpu_up_read(&sbi->s_journal_flag_rwsem);
+	percpu_up_read(&sbi->s_writepages_rwsem);
 	return ret;
 }
 
@@ -5861,7 +5861,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 		}
 	}
 
-	percpu_down_write(&sbi->s_journal_flag_rwsem);
+	percpu_down_write(&sbi->s_writepages_rwsem);
 	jbd2_journal_lock_updates(journal);
 
 	/*
@@ -5878,15 +5878,15 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 		err = jbd2_journal_flush(journal);
 		if (err < 0) {
 			jbd2_journal_unlock_updates(journal);
-			percpu_up_write(&sbi->s_journal_flag_rwsem);
+			percpu_up_write(&sbi->s_writepages_rwsem);
 			return err;
 		}
 		ext4_clear_inode_flag(inode, EXT4_INODE_JOURNAL_DATA);
 	}
 	ext4_set_aops(inode);
 
 	jbd2_journal_unlock_updates(journal);
-	percpu_up_write(&sbi->s_journal_flag_rwsem);
+	percpu_up_write(&sbi->s_writepages_rwsem);
 
 	if (val)
 		up_write(&EXT4_I(inode)->i_mmap_sem);