diff mbox series

[v2,01/11] ext4 resise: extra brelse in setup_new_flex_group_blocks()

Message ID 37de5906-2d76-461a-ff92-b6fcc775b408@virtuozzo.com
State Accepted, archived
Headers show
Series ext4: number of long lived errors | expand

Commit Message

Vasily Averin Oct. 30, 2018, 9:57 p.m. UTC 
currently bh is set to NULL only during first iteration of for cycle,
then this pointer is not cleared after end of using.
Therefore rollback after errors can lead to extra brelse(bh) call,
decrements bh counter and later trigger an unexpected warning in __brelse()

Patch moves brelse() calls in body of cycle to exclude requirement of
brelse() call in rollback.

Fixes 33afdcc5402d ("ext4: add a function which sets up group blocks ...") # 3.3+

Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
---
 fs/ext4/resize.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

Comments

Theodore Ts'o Nov. 3, 2018, 8:20 p.m. UTC | #1 
On Wed, Oct 31, 2018 at 12:57:37AM +0300, Vasily Averin wrote:
> currently bh is set to NULL only during first iteration of for cycle,
> then this pointer is not cleared after end of using.
> Therefore rollback after errors can lead to extra brelse(bh) call,
> decrements bh counter and later trigger an unexpected warning in __brelse()
> 
> Patch moves brelse() calls in body of cycle to exclude requirement of
> brelse() call in rollback.
> 
> Fixes 33afdcc5402d ("ext4: add a function which sets up group blocks ...") # 3.3+
> 
> Signed-off-by: Vasily Averin <vvs@virtuozzo.com>

Thanks, applied.  I addjusted the patch summary to read:

    ext4: avoid potential extra brelse in setup_new_flex_group_blocks()

(Note that resise should have been spelled as "resize", by the way.)


						- Ted
diff mbox series

Patch

diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index ebbc663d0798..c3fa30878ca8 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -605,7 +605,6 @@  static int setup_new_flex_group_blocks(struct super_block *sb,
 		bh = bclean(handle, sb, block);
 		if (IS_ERR(bh)) {
 			err = PTR_ERR(bh);
-			bh = NULL;
 			goto out;
 		}
 		overhead = ext4_group_overhead_blocks(sb, group);
@@ -618,9 +617,9 @@  static int setup_new_flex_group_blocks(struct super_block *sb,
 		ext4_mark_bitmap_end(EXT4_B2C(sbi, group_data[i].blocks_count),
 				     sb->s_blocksize * 8, bh->b_data);
 		err = ext4_handle_dirty_metadata(handle, NULL, bh);
+		brelse(bh);
 		if (err)
 			goto out;
-		brelse(bh);
 
 handle_ib:
 		if (bg_flags[i] & EXT4_BG_INODE_UNINIT)
@@ -635,18 +634,16 @@  static int setup_new_flex_group_blocks(struct super_block *sb,
 		bh = bclean(handle, sb, block);
 		if (IS_ERR(bh)) {
 			err = PTR_ERR(bh);
-			bh = NULL;
 			goto out;
 		}
 
 		ext4_mark_bitmap_end(EXT4_INODES_PER_GROUP(sb),
 				     sb->s_blocksize * 8, bh->b_data);
 		err = ext4_handle_dirty_metadata(handle, NULL, bh);
+		brelse(bh);
 		if (err)
 			goto out;
-		brelse(bh);
 	}
-	bh = NULL;
 
 	/* Mark group tables in block bitmap */
 	for (j = 0; j < GROUP_TABLE_COUNT; j++) {
@@ -685,7 +682,6 @@  static int setup_new_flex_group_blocks(struct super_block *sb,
 	}
 
 out:
-	brelse(bh);
 	err2 = ext4_journal_stop(handle);
 	if (err2 && !err)
 		err = err2;