f2fs: fix writing incorrect orphan blocks

Previously, there was a erroneous scenario like below.
thread 1:                       thread 2:
 f2fs_unlink
  - acquire_orphan_inode
    : sbi->n_orphans++           write_checkpoint
                                 - block_operations
                                  : f2fs_lock_all
                                 - do_checkpoint
                                  : write orphan blocks with sbi->n_orphans
                                 - unblock_operations
  - f2fs_lock_op
  - release_orphan_inode
  - f2fs_unlock_op

During the checkpoint by thread 2, f2fs stores a wrong orphan block according
to the wrong sbi->n_orphans.
To avoid this, simply we should make cover acquire_orphan_inode too with
f2fs_lock_op.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This commit is contained in:
Jaegeuk Kim 2013-10-08 10:19:28 +09:00
parent 5887d291d7
commit ccaaca2591
1 changed files with 2 additions and 2 deletions

View File

@ -228,14 +228,14 @@ static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
if (!de) if (!de)
goto fail; goto fail;
f2fs_lock_op(sbi);
err = acquire_orphan_inode(sbi); err = acquire_orphan_inode(sbi);
if (err) { if (err) {
f2fs_unlock_op(sbi);
kunmap(page); kunmap(page);
f2fs_put_page(page, 0); f2fs_put_page(page, 0);
goto fail; goto fail;
} }
f2fs_lock_op(sbi);
f2fs_delete_entry(de, page, inode); f2fs_delete_entry(de, page, inode);
f2fs_unlock_op(sbi); f2fs_unlock_op(sbi);