ocfs2: Re-order iput in ocfs2_drop_dentry_lock
authorMark Fasheh <mark.fasheh@oracle.com>
Thu, 18 Oct 2007 19:36:10 +0000 (12:36 -0700)
committerMark Fasheh <mark.fasheh@oracle.com>
Tue, 6 Nov 2007 23:31:52 +0000 (15:31 -0800)
commit9f70968af3e6e21612e06e153aa71c62dee5a09b
treec4a11e07d94ad99fe13512b3b1a117de2613c300
parent019d1b2247c6898589560c6f3b3e7ec280b0010a
ocfs2: Re-order iput in ocfs2_drop_dentry_lock

Do this to avoid a theoretical (I haven't seen this in practice) race where
the downconvert thread might drop the dentry lock, allowing a remote unlink
to proceed before dropping the inode locks. This could bounce access to the
orphan dir between nodes.

There doesn't seem to be a need to do the same in ocfs2_dentry_iput() as
that's never called for the last ref drop from the downconvert thread.

Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
fs/ocfs2/dcache.c