]> git.ipfire.org Git - thirdparty/kernel/stable.git/commitdiff
ocfs2: prevent release journal inode after journal shutdown
authorEdward Adam Davis <eadavis@qq.com>
Tue, 19 Aug 2025 13:41:02 +0000 (21:41 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 9 Sep 2025 16:58:16 +0000 (18:58 +0200)
commit f46e8ef8bb7b452584f2e75337b619ac51a7cadf upstream.

Before calling ocfs2_delete_osb(), ocfs2_journal_shutdown() has already
been executed in ocfs2_dismount_volume(), so osb->journal must be NULL.
Therefore, the following calltrace will inevitably fail when it reaches
jbd2_journal_release_jbd_inode().

ocfs2_dismount_volume()->
  ocfs2_delete_osb()->
    ocfs2_free_slot_info()->
      __ocfs2_free_slot_info()->
        evict()->
          ocfs2_evict_inode()->
            ocfs2_clear_inode()->
      jbd2_journal_release_jbd_inode(osb->journal->j_journal,

Adding osb->journal checks will prevent null-ptr-deref during the above
execution path.

Link: https://lkml.kernel.org/r/tencent_357489BEAEE4AED74CBD67D246DBD2C4C606@qq.com
Fixes: da5e7c87827e ("ocfs2: cleanup journal init and shutdown")
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
Reported-by: syzbot+47d8cb2f2cc1517e515a@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=47d8cb2f2cc1517e515a
Tested-by: syzbot+47d8cb2f2cc1517e515a@syzkaller.appspotmail.com
Reviewed-by: Mark Tinguely <mark.tinguely@oracle.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Jun Piao <piaojun@huawei.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs/ocfs2/inode.c

index 2cc5c99fe94167bd5c01e7fd050651ba4772a76d..4a7509389cf38d924c6be815ae9b6ce67c774c62 100644 (file)
@@ -1205,6 +1205,9 @@ static void ocfs2_clear_inode(struct inode *inode)
         * the journal is flushed before journal shutdown. Thus it is safe to
         * have inodes get cleaned up after journal shutdown.
         */
+       if (!osb->journal)
+               return;
+
        jbd2_journal_release_jbd_inode(osb->journal->j_journal,
                                       &oi->ip_jinode);
 }